feat: add training and evaluation logs for Qwen models
Browse filesIntroduced new log files for various training and evaluation runs, including SFT and GRPO tasks for Qwen 2.5-7B and DeepSeek 14B models. Enhanced logging details to capture command execution, runtime metrics, and evaluation results, facilitating better tracking and analysis of model performance during training and evaluation processes.
- training_logs/1354.txt +414 -0
- training_logs/221184.txt +48 -0
- training_logs/288278.txt +25 -0
- training_logs/309711.txt +155 -0
- training_logs/386081.txt +47 -0
- training_logs/432673.txt +26 -0
- training_logs/524754.txt +207 -0
- training_logs/546061.txt +25 -0
- training_logs/829019.txt +105 -0
training_logs/1354.txt
ADDED
|
@@ -0,0 +1,414 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 0 |
0%| | 0/150 [00:00<?, ?it/s]Passing `generation_config` together with generation-related arguments=({'cache_implementation', 'disable_compile', 'pad_token_id'}) is deprecated and will be removed in future versions. Please pass either a `generation_config` object OR all generation parameters explicitly, but not both.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
1%| | 1/150 [00:41<1:43:23, 41.64s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 2 |
1%|β | 2/150 [01:08<1:21:40, 33.11s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 3 |
2%|β | 3/150 [01:55<1:36:41, 39.46s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 4 |
3%|β | 4/150 [02:42<1:42:40, 42.20s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 5 |
3%|β | 5/150 [03:00<1:20:50, 33.45s/it]
|
| 6 |
|
|
|
|
| 7 |
3%|β | 5/150 [03:00<1:20:50, 33.45s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 8 |
4%|β | 6/150 [03:30<1:17:51, 32.44s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 9 |
5%|β | 7/150 [04:02<1:16:34, 32.13s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 10 |
5%|β | 8/150 [04:51<1:29:21, 37.76s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 11 |
6%|β | 9/150 [05:40<1:36:46, 41.18s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 12 |
7%|β | 10/150 [06:29<1:41:47, 43.63s/it]
|
| 13 |
|
|
|
|
| 14 |
7%|β | 10/150 [06:29<1:41:47, 43.63s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 15 |
7%|β | 11/150 [07:00<1:31:51, 39.65s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 16 |
8%|β | 12/150 [07:28<1:23:12, 36.18s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 17 |
9%|β | 13/150 [08:17<1:31:16, 39.98s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 18 |
9%|β | 14/150 [08:38<1:17:25, 34.16s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 19 |
10%|β | 15/150 [09:27<1:27:10, 38.74s/it]
|
| 20 |
|
|
|
|
| 21 |
10%|β | 15/150 [09:27<1:27:10, 38.74s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 22 |
11%|β | 16/150 [10:02<1:23:47, 37.52s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 23 |
11%|ββ | 17/150 [10:33<1:19:19, 35.79s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 24 |
12%|ββ | 18/150 [11:02<1:14:05, 33.68s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 25 |
13%|ββ | 19/150 [11:51<1:23:25, 38.21s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 26 |
13%|ββ | 20/150 [12:17<1:14:37, 34.44s/it]
|
| 27 |
|
|
|
|
| 28 |
13%|ββ | 20/150 [12:17<1:14:37, 34.44s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 29 |
14%|ββ | 21/150 [13:05<1:23:13, 38.71s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 30 |
15%|ββ | 22/150 [13:54<1:29:01, 41.73s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 31 |
15%|ββ | 23/150 [14:19<1:17:39, 36.69s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 32 |
16%|ββ | 24/150 [15:07<1:24:19, 40.15s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 33 |
17%|ββ | 25/150 [15:27<1:11:10, 34.16s/it]
|
| 34 |
|
|
|
|
| 35 |
17%|ββ | 25/150 [15:27<1:11:10, 34.16s/it]Unsloth: Restored added_tokens_decoder metadata in outputs/grpo_adapter/checkpoint-25/tokenizer_config.json.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 36 |
17%|ββ | 26/150 [16:05<1:13:02, 35.34s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 37 |
18%|ββ | 27/150 [16:37<1:10:06, 34.20s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 38 |
19%|ββ | 28/150 [17:27<1:19:02, 38.87s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 39 |
19%|ββ | 29/150 [17:56<1:12:30, 35.95s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 40 |
20%|ββ | 30/150 [18:24<1:07:18, 33.65s/it]
|
| 41 |
|
|
|
|
| 42 |
20%|ββ | 30/150 [18:24<1:07:18, 33.65s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 43 |
21%|ββ | 31/150 [18:56<1:05:26, 32.99s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 44 |
21%|βββ | 32/150 [19:28<1:04:17, 32.69s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 45 |
22%|βββ | 33/150 [19:44<53:58, 27.68s/it] Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 46 |
23%|βββ | 34/150 [20:24<1:00:42, 31.40s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 47 |
23%|βββ | 35/150 [21:14<1:11:16, 37.19s/it]
|
| 48 |
|
|
|
|
| 49 |
23%|βββ | 35/150 [21:14<1:11:16, 37.19s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 50 |
24%|βββ | 36/150 [21:49<1:09:00, 36.32s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 51 |
25%|βββ | 37/150 [22:39<1:16:10, 40.44s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 52 |
25%|βββ | 38/150 [23:04<1:07:08, 35.97s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 53 |
26%|βββ | 39/150 [23:42<1:07:27, 36.46s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 54 |
27%|βββ | 40/150 [24:33<1:15:08, 40.98s/it]
|
| 55 |
|
|
|
|
| 56 |
27%|βββ | 40/150 [24:33<1:15:08, 40.98s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 57 |
27%|βββ | 41/150 [25:00<1:06:32, 36.63s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 58 |
28%|βββ | 42/150 [25:18<56:04, 31.16s/it] Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 59 |
29%|βββ | 43/150 [25:45<53:06, 29.78s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 60 |
29%|βββ | 44/150 [26:19<55:08, 31.21s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 61 |
30%|βββ | 45/150 [27:08<1:03:36, 36.35s/it]
|
| 62 |
|
|
|
|
| 63 |
30%|βββ | 45/150 [27:08<1:03:36, 36.35s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 64 |
31%|βββ | 46/150 [27:57<1:09:50, 40.30s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 65 |
31%|ββββ | 47/150 [28:39<1:10:07, 40.85s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 66 |
32%|ββββ | 48/150 [29:28<1:13:14, 43.09s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 67 |
33%|ββββ | 49/150 [30:16<1:15:10, 44.66s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 68 |
33%|ββββ | 50/150 [30:42<1:04:53, 38.93s/it]
|
| 69 |
|
|
|
|
| 70 |
33%|ββββ | 50/150 [30:42<1:04:53, 38.93s/it]Unsloth: Restored added_tokens_decoder metadata in outputs/grpo_adapter/checkpoint-50/tokenizer_config.json.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 71 |
34%|ββββ | 51/150 [31:11<59:43, 36.20s/it] Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 72 |
35%|ββββ | 52/150 [31:40<55:17, 33.85s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 73 |
35%|ββββ | 53/150 [32:05<50:19, 31.13s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 74 |
36%|ββββ | 54/150 [32:54<58:45, 36.72s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 75 |
37%|ββββ | 55/150 [33:43<1:03:55, 40.38s/it]
|
| 76 |
|
|
|
|
| 77 |
37%|ββββ | 55/150 [33:43<1:03:55, 40.38s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 78 |
37%|ββββ | 56/150 [34:14<58:58, 37.64s/it] Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 79 |
38%|ββββ | 57/150 [34:38<51:44, 33.38s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 80 |
39%|ββββ | 58/150 [35:27<58:16, 38.01s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 81 |
39%|ββββ | 59/150 [35:58<54:42, 36.07s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 82 |
40%|ββββ | 60/150 [36:30<51:58, 34.65s/it]
|
| 83 |
|
|
|
|
| 84 |
40%|ββββ | 60/150 [36:30<51:58, 34.65s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 85 |
41%|ββββ | 61/150 [37:19<57:47, 38.96s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 86 |
41%|βββββ | 62/150 [38:08<1:01:50, 42.17s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 87 |
42%|βββββ | 63/150 [38:31<52:35, 36.27s/it] Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 88 |
43%|βββββ | 64/150 [38:59<48:25, 33.79s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 89 |
43%|βββββ | 65/150 [39:21<43:05, 30.42s/it]
|
| 90 |
|
|
|
|
| 91 |
43%|βββββ | 65/150 [39:21<43:05, 30.42s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 92 |
44%|βββββ | 66/150 [40:10<50:23, 36.00s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 93 |
45%|βββββ | 67/150 [40:59<55:14, 39.93s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 94 |
45%|βββββ | 68/150 [41:31<51:02, 37.35s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 95 |
46%|βββββ | 69/150 [42:19<54:59, 40.74s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 96 |
47%|βββββ | 70/150 [43:03<55:21, 41.52s/it]
|
| 97 |
|
|
|
|
| 98 |
47%|βββββ | 70/150 [43:03<55:21, 41.52s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 99 |
47%|βββββ | 71/150 [43:34<50:33, 38.40s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 100 |
48%|βββββ | 72/150 [44:23<54:00, 41.55s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 101 |
49%|βββββ | 73/150 [44:48<46:52, 36.53s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 102 |
49%|βββββ | 74/150 [45:26<46:53, 37.02s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 103 |
50%|βββββ | 75/150 [45:48<40:55, 32.74s/it]
|
| 104 |
|
|
|
|
| 105 |
50%|βββββ | 75/150 [45:48<40:55, 32.74s/it]Unsloth: Restored added_tokens_decoder metadata in outputs/grpo_adapter/checkpoint-75/tokenizer_config.json.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 106 |
51%|βββββ | 76/150 [46:14<37:39, 30.53s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 107 |
51%|ββββββ | 77/150 [47:03<43:48, 36.01s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 108 |
52%|ββββββ | 78/150 [47:38<42:48, 35.67s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 109 |
53%|ββββββ | 79/150 [48:29<47:49, 40.41s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 110 |
53%|ββββββ | 80/150 [49:17<49:54, 42.78s/it]
|
| 111 |
|
|
|
|
| 112 |
53%|ββββββ | 80/150 [49:17<49:54, 42.78s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 113 |
54%|ββββββ | 81/150 [49:49<45:15, 39.36s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 114 |
55%|ββββββ | 82/150 [50:15<40:14, 35.50s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 115 |
55%|ββββββ | 83/150 [50:48<38:35, 34.56s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 116 |
56%|ββββββ | 84/150 [51:36<42:34, 38.70s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 117 |
57%|ββββββ | 85/150 [52:07<39:26, 36.41s/it]
|
| 118 |
|
|
|
|
| 119 |
57%|ββββββ | 85/150 [52:07<39:26, 36.41s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 120 |
57%|ββββββ | 86/150 [52:33<35:23, 33.18s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 121 |
58%|ββββββ | 87/150 [53:07<35:14, 33.56s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 122 |
59%|ββββββ | 88/150 [53:29<31:01, 30.03s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 123 |
59%|ββββββ | 89/150 [54:10<33:57, 33.39s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 124 |
60%|ββββββ | 90/150 [54:38<31:37, 31.63s/it]
|
| 125 |
|
|
|
|
| 126 |
60%|ββββββ | 90/150 [54:38<31:37, 31.63s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 127 |
61%|ββββββ | 91/150 [55:06<30:02, 30.55s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 128 |
61%|βββββββ | 92/150 [55:33<28:28, 29.46s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 129 |
62%|βββββββ | 93/150 [56:23<33:49, 35.61s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 130 |
63%|βββββββ | 94/150 [56:53<31:55, 34.21s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 131 |
63%|βββββββ | 95/150 [57:19<29:00, 31.64s/it]
|
| 132 |
|
|
|
|
| 133 |
63%|βββββββ | 95/150 [57:19<29:00, 31.64s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 134 |
64%|βββββββ | 96/150 [57:55<29:34, 32.87s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 135 |
65%|βββββββ | 97/150 [58:43<32:58, 37.33s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 136 |
65%|βββββββ | 98/150 [59:17<31:36, 36.48s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 137 |
66%|βββββββ | 99/150 [1:00:06<34:10, 40.20s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 138 |
67%|βββββββ | 100/150 [1:00:48<34:02, 40.85s/it]
|
| 139 |
|
|
|
|
| 140 |
67%|βββββββ | 100/150 [1:00:48<34:02, 40.85s/it]Unsloth: Restored added_tokens_decoder metadata in outputs/grpo_adapter/checkpoint-100/tokenizer_config.json.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 141 |
67%|βββββββ | 101/150 [1:01:12<29:02, 35.56s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 142 |
68%|βββββββ | 102/150 [1:02:00<31:34, 39.47s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 143 |
69%|βββββββ | 103/150 [1:02:28<28:07, 35.91s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 144 |
69%|βββββββ | 104/150 [1:03:02<27:08, 35.41s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 145 |
70%|βββββββ | 105/150 [1:03:51<29:34, 39.43s/it]
|
| 146 |
|
|
|
|
| 147 |
70%|βββββββ | 105/150 [1:03:51<29:34, 39.43s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 148 |
71%|βββββββ | 106/150 [1:04:39<30:55, 42.18s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 149 |
71%|ββββββββ | 107/150 [1:05:27<31:29, 43.93s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 150 |
72%|ββββββββ | 108/150 [1:05:47<25:35, 36.57s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 151 |
73%|ββββββββ | 109/150 [1:06:32<26:51, 39.30s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 152 |
73%|ββββββββ | 110/150 [1:07:01<23:56, 35.92s/it]
|
| 153 |
|
|
|
|
| 154 |
73%|ββββββββ | 110/150 [1:07:01<23:56, 35.92s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 155 |
74%|ββββββββ | 111/150 [1:07:27<21:36, 33.23s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 156 |
75%|ββββββββ | 112/150 [1:07:53<19:29, 30.77s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 157 |
75%|ββββββββ | 113/150 [1:08:23<19:00, 30.82s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 158 |
76%|ββββββββ | 114/150 [1:09:12<21:40, 36.13s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 159 |
77%|ββββββββ | 115/150 [1:09:51<21:36, 37.05s/it]
|
| 160 |
|
|
|
|
| 161 |
77%|ββββββββ | 115/150 [1:09:51<21:36, 37.05s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 162 |
77%|ββββββββ | 116/150 [1:10:22<20:01, 35.34s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 163 |
78%|ββββββββ | 117/150 [1:11:19<22:54, 41.66s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 164 |
79%|ββββββββ | 118/150 [1:11:39<18:44, 35.16s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 165 |
79%|ββββββββ | 119/150 [1:12:27<20:09, 39.03s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 166 |
80%|ββββββββ | 120/150 [1:12:56<17:56, 35.89s/it]
|
| 167 |
|
|
|
|
| 168 |
80%|ββββββββ | 120/150 [1:12:56<17:56, 35.89s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 169 |
81%|ββββββββ | 121/150 [1:13:53<20:28, 42.37s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 170 |
81%|βββββββββ | 122/150 [1:14:42<20:40, 44.31s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 171 |
82%|βββββββββ | 123/150 [1:15:22<19:21, 43.01s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 172 |
83%|βββββββββ | 124/150 [1:15:57<17:36, 40.62s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 173 |
83%|βββββββββ | 125/150 [1:16:46<17:56, 43.06s/it]
|
| 174 |
|
|
|
|
| 175 |
83%|βββββββββ | 125/150 [1:16:46<17:56, 43.06s/it]Unsloth: Restored added_tokens_decoder metadata in outputs/grpo_adapter/checkpoint-125/tokenizer_config.json.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 176 |
84%|βββββββββ | 126/150 [1:17:36<18:04, 45.19s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 177 |
85%|βββββββββ | 127/150 [1:18:10<16:02, 41.86s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 178 |
85%|βββββββββ | 128/150 [1:18:36<13:39, 37.24s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 179 |
86%|βββββββββ | 129/150 [1:19:26<14:22, 41.05s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 180 |
87%|βββββββββ | 130/150 [1:19:48<11:44, 35.24s/it]
|
| 181 |
|
|
|
|
| 182 |
87%|βββββββββ | 130/150 [1:19:48<11:44, 35.24s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 183 |
87%|βββββββββ | 131/150 [1:20:37<12:26, 39.31s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 184 |
88%|βββββββββ | 132/150 [1:21:08<11:03, 36.86s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 185 |
89%|βββββββββ | 133/150 [1:21:34<09:31, 33.63s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 186 |
89%|βββββββββ | 134/150 [1:21:59<08:15, 30.99s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 187 |
90%|βββββββββ | 135/150 [1:22:35<08:07, 32.50s/it]
|
| 188 |
|
|
|
|
| 189 |
90%|βββββββββ | 135/150 [1:22:35<08:07, 32.50s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 190 |
91%|βββββββββ | 136/150 [1:23:07<07:34, 32.49s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 191 |
91%|ββββββββββ| 137/150 [1:23:56<08:05, 37.35s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 192 |
92%|ββββββββββ| 138/150 [1:24:25<06:56, 34.73s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 193 |
93%|ββββββββββ| 139/150 [1:25:05<06:40, 36.39s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 194 |
93%|ββββββββββ| 140/150 [1:25:40<05:59, 35.92s/it]
|
| 195 |
|
|
|
|
| 196 |
93%|ββββββββββ| 140/150 [1:25:40<05:59, 35.92s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 197 |
94%|ββββββββββ| 141/150 [1:26:00<04:41, 31.23s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 198 |
95%|ββββββββββ| 142/150 [1:26:24<03:51, 28.91s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 199 |
95%|ββββββββββ| 143/150 [1:26:43<03:01, 25.95s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 200 |
96%|ββββββββββ| 144/150 [1:27:16<02:49, 28.33s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 201 |
97%|ββββββββββ| 145/150 [1:28:05<02:51, 34.29s/it]
|
| 202 |
|
|
|
|
| 203 |
97%|ββββββββββ| 145/150 [1:28:05<02:51, 34.29s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 204 |
97%|ββββββββββ| 146/150 [1:28:35<02:12, 33.15s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 205 |
98%|ββββββββββ| 147/150 [1:29:07<01:38, 32.68s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 206 |
99%|ββββββββββ| 148/150 [1:29:32<01:00, 30.38s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 207 |
99%|ββββββββββ| 149/150 [1:30:21<00:35, 35.90s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
|
|
|
| 208 |
|
|
|
|
|
|
|
| 209 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 210 |
...adapter/training_args.bin: 100%|ββββββββββ| 6.74kB / 6.74kB [A[A
|
|
|
|
|
|
|
|
|
|
| 211 |
...po_adapter/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB [A[A[A
|
|
|
|
|
|
|
|
|
|
|
|
|
| 212 |
...adapter_model.safetensors: 100%|ββββββββββ| 162MB / 162MB [A[A[A[A
|
|
|
|
|
|
|
| 213 |
...adapter/training_args.bin: 100%|ββββββββββ| 6.74kB / 6.74kB [A[A
|
|
|
|
|
|
|
|
|
|
| 214 |
...po_adapter/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB [A[A[A
|
|
|
|
|
|
|
|
|
|
|
|
|
| 215 |
...adapter_model.safetensors: 100%|ββββββββββ| 162MB / 162MB [A[A[A[A
|
|
|
|
|
|
|
| 216 |
...adapter/training_args.bin: 100%|ββββββββββ| 6.74kB / 6.74kB [A[A
|
|
|
|
|
|
|
|
|
|
| 217 |
...po_adapter/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB [A[A[A
|
|
|
|
|
|
|
|
|
|
|
|
|
| 218 |
...adapter_model.safetensors: 100%|ββββββββββ| 162MB / 162MB [A[A[A[A
|
|
|
|
|
|
|
| 219 |
...adapter/training_args.bin: 100%|ββββββββββ| 6.74kB / 6.74kB [A[A
|
|
|
|
|
|
|
|
|
|
| 220 |
...po_adapter/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB [A[A[A
|
|
|
|
|
|
|
|
|
|
|
|
|
| 221 |
...adapter_model.safetensors: 100%|ββββββββββ| 162MB / 162MB [A[A[A[A
|
|
|
|
|
|
|
| 222 |
...adapter/training_args.bin: 100%|ββββββββββ| 6.74kB / 6.74kB [A[A
|
|
|
|
|
|
|
|
|
|
| 223 |
...po_adapter/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB [A[A[A
|
|
|
|
|
|
|
|
|
|
|
|
|
| 224 |
...adapter_model.safetensors: 100%|ββββββββββ| 162MB / 162MB [A[A[A[A
|
|
|
|
|
|
|
| 225 |
...adapter/training_args.bin: 100%|ββββββββββ| 6.74kB / 6.74kB
|
|
|
|
| 226 |
...po_adapter/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB
|
|
|
|
| 227 |
...adapter_model.safetensors: 100%|ββββββββββ| 162MB / 162MB
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 228 |
...adapter/training_args.bin: 100%|ββββββββββ| 6.74kB / 6.74kB [A[A
|
|
|
|
|
|
|
|
|
|
| 229 |
...po_adapter/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB [A[A[A
|
|
|
|
|
|
|
|
|
|
|
|
|
| 230 |
...adapter_model.safetensors: 100%|ββββββββββ| 162MB / 162MB [A[A[A[A
|
|
|
|
|
|
|
| 231 |
...adapter/training_args.bin: 100%|ββββββββββ| 6.74kB / 6.74kB [A[A
|
|
|
|
|
|
|
|
|
|
| 232 |
...po_adapter/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB [A[A[A
|
|
|
|
|
|
|
|
|
|
|
|
|
| 233 |
...adapter_model.safetensors: 100%|ββββββββββ| 162MB / 162MB [A[A[A[A
|
|
|
|
|
|
|
| 234 |
...adapter/training_args.bin: 100%|ββββββββββ| 6.74kB / 6.74kB [A[A
|
|
|
|
|
|
|
|
|
|
| 235 |
...po_adapter/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB [A[A[A
|
|
|
|
|
|
|
|
|
|
|
|
|
| 236 |
...adapter_model.safetensors: 100%|ββββββββββ| 162MB / 162MB [A[A[A[A
|
|
|
|
|
|
|
| 237 |
...adapter/training_args.bin: 100%|ββββββββββ| 6.74kB / 6.74kB [A[A
|
|
|
|
|
|
|
|
|
|
| 238 |
...po_adapter/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB [A[A[A
|
|
|
|
|
|
|
|
|
|
|
|
|
| 239 |
...adapter_model.safetensors: 100%|ββββββββββ| 162MB / 162MB [A[A[A[A
|
|
|
|
|
|
|
| 240 |
...adapter/training_args.bin: 100%|ββββββββββ| 6.74kB / 6.74kB
|
|
|
|
| 241 |
...po_adapter/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB
|
|
|
|
| 242 |
...adapter_model.safetensors: 100%|ββββββββββ| 162MB / 162MB
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
pid: 125784
|
| 3 |
+
cwd: "/teamspace/studios/this_studio/corp_gym"
|
| 4 |
+
command: "stdbuf -oL -eL python -u training/train_grpo.py \\\n --model Qwen/Qwen2.5-7B-Instruct \\\n --adapter outputs/sft_adapter \\\n --examples data/processed/e1_m1_clean.jsonl,data/processed/h1_seed_clean.jsonl \\\n --tasks e1_launch_readiness,m1_budget_reallocation,h1_acquisition_defence \\\n --output outputs/grpo_adapter \\\n --max-steps 150 \\\n --generations 2 \\\n --batch-size 1 --grad-accum 8 \\\n --dataloader-num-workers 2 \\\n --push-to-hub Navigam/corp-env-grpo-qwen2.5-7b 2>&1 | tee results/runs/grpo_train.log"
|
| 5 |
+
started_at: 2026-04-25T16:48:36.348Z
|
| 6 |
+
running_for_ms: 5490099
|
| 7 |
+
---
|
| 8 |
+
π¦₯ Unsloth: Will patch your computer to enable 2x faster free finetuning.
|
| 9 |
+
π¦₯ Unsloth Zoo will now patch everything to make training faster!
|
| 10 |
+
Unsloth: UnslothBCOTrainer is already patched.
|
| 11 |
+
Unsloth: UnslothCPOTrainer is already patched.
|
| 12 |
+
Unsloth: UnslothDPOTrainer is already patched.
|
| 13 |
+
Unsloth: UnslothGKDTrainer is already patched.
|
| 14 |
+
Unsloth: UnslothGRPOTrainer is already patched.
|
| 15 |
+
Unsloth: UnslothKTOTrainer is already patched.
|
| 16 |
+
Unsloth: UnslothNashMDTrainer is already patched.
|
| 17 |
+
Unsloth: UnslothOnlineDPOTrainer is already patched.
|
| 18 |
+
Unsloth: UnslothORPOTrainer is already patched.
|
| 19 |
+
Unsloth: UnslothPPOTrainer is already patched.
|
| 20 |
+
Unsloth: UnslothPRMTrainer is already patched.
|
| 21 |
+
Unsloth: UnslothRewardTrainer is already patched.
|
| 22 |
+
Unsloth: UnslothRLOOTrainer is already patched.
|
| 23 |
+
Unsloth: UnslothSFTTrainer is already patched.
|
| 24 |
+
Unsloth: UnslothXPOTrainer is already patched.
|
| 25 |
+
==((====))== Unsloth 2026.4.8: Fast Qwen2 patching. Transformers: 5.5.0.
|
| 26 |
+
\\ /| NVIDIA H100 80GB HBM3. Num GPUs = 1. Max memory: 79.189 GB. Platform: Linux.
|
| 27 |
+
O^O/ \_/ \ Torch: 2.7.1+cu128. CUDA: 9.0. CUDA Toolkit: 12.8. Triton: 3.3.1
|
| 28 |
+
\ / Bfloat16 = TRUE. FA [Xformers = 0.0.31.post1. FA2 = True]
|
| 29 |
+
"-____-" Free license: http://github.com/unslothai/unsloth
|
| 30 |
+
Unsloth: Fast downloading is enabled - ignore downloading bars which are red colored!
|
| 31 |
+
|
| 32 |
+
unsloth/qwen2.5-7b-instruct-unsloth-bnb-4bit does not have a padding token! Will use pad_token = <|PAD_TOKEN|>.
|
| 33 |
+
GRPO dataset: kept 1143 / 1384 prompts (<= 7372 tokens); skipped 241 oversized prompts.
|
| 34 |
+
The tokenizer has new PAD/BOS/EOS tokens that differ from the model config and generation config. The model config and generation config were aligned accordingly, being updated with the tokenizer's values. Updated tokens: {'bos_token_id': None}.
|
| 35 |
+
==((====))== Unsloth - 2x faster free finetuning | Num GPUs used = 1
|
| 36 |
+
\\ /| Num examples = 1,143 | Num Epochs = 1 | Total steps = 150
|
| 37 |
+
O^O/ \_/ \ Batch size per device = 1 | Gradient accumulation steps = 8
|
| 38 |
+
\ / Data Parallel GPUs = 1 | Total batch size (1 x 8 x 1) = 8
|
| 39 |
+
"-____-" Trainable parameters = 80,740,352 of 7,696,356,864 (1.05% trained)
|
| 40 |
+
|
| 41 |
0%| | 0/150 [00:00<?, ?it/s]Passing `generation_config` together with generation-related arguments=({'cache_implementation', 'disable_compile', 'pad_token_id'}) is deprecated and will be removed in future versions. Please pass either a `generation_config` object OR all generation parameters explicitly, but not both.
|
| 42 |
+
Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 43 |
+
/home/zeus/miniconda3/envs/cloudspace/lib/python3.12/site-packages/transformers/modeling_attn_mask_utils.py:71: FutureWarning: The attention mask API under `transformers.modeling_attn_mask_utils` (`AttentionMaskConverter`) is deprecated and will be removed in Transformers v5.10. Please use the new API in `transformers.masking_utils`.
|
| 44 |
+
warnings.warn(DEPRECATION_MESSAGE, FutureWarning)
|
| 45 |
+
/home/zeus/miniconda3/envs/cloudspace/lib/python3.12/site-packages/transformers/modeling_attn_mask_utils.py:281: FutureWarning: The attention mask API under `transformers.modeling_attn_mask_utils` (`AttentionMaskConverter`) is deprecated and will be removed in Transformers v5.10. Please use the new API in `transformers.masking_utils`.
|
| 46 |
+
warnings.warn(DEPRECATION_MESSAGE, FutureWarning)
|
| 47 |
+
/home/zeus/miniconda3/envs/cloudspace/lib/python3.12/site-packages/transformers/modeling_attn_mask_utils.py:172: FutureWarning: The attention mask API under `transformers.modeling_attn_mask_utils` (`AttentionMaskConverter`) is deprecated and will be removed in Transformers v5.10. Please use the new API in `transformers.masking_utils`.
|
| 48 |
+
warnings.warn(DEPRECATION_MESSAGE, FutureWarning)
|
| 49 |
+
/home/zeus/miniconda3/envs/cloudspace/lib/python3.12/site-packages/transformers/modeling_attn_mask_utils.py:202: FutureWarning: The attention mask API under `transformers.modeling_attn_mask_utils` (`AttentionMaskConverter`) is deprecated and will be removed in Transformers v5.10. Please use the new API in `transformers.masking_utils`.
|
| 50 |
+
warnings.warn(DEPRECATION_MESSAGE, FutureWarning)
|
| 51 |
+
/home/zeus/miniconda3/envs/cloudspace/lib/python3.12/site-packages/transformers/modeling_attn_mask_utils.py:254: FutureWarning: The attention mask API under `transformers.modeling_attn_mask_utils` (`AttentionMaskConverter`) is deprecated and will be removed in Transformers v5.10. Please use the new API in `transformers.masking_utils`.
|
| 52 |
+
warnings.warn(DEPRECATION_MESSAGE, FutureWarning)
|
| 53 |
+
`use_return_dict` is deprecated! Use `return_dict` instead!
|
| 54 |
+
Unsloth: Will smartly offload gradients to save VRAM!
|
| 55 |
+
|
| 56 |
1%| | 1/150 [00:41<1:43:23, 41.64s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 57 |
+
/home/zeus/miniconda3/envs/cloudspace/lib/python3.12/site-packages/transformers/modeling_attn_mask_utils.py:71: FutureWarning: The attention mask API under `transformers.modeling_attn_mask_utils` (`AttentionMaskConverter`) is deprecated and will be removed in Transformers v5.10. Please use the new API in `transformers.masking_utils`.
|
| 58 |
+
warnings.warn(DEPRECATION_MESSAGE, FutureWarning)
|
| 59 |
+
/home/zeus/miniconda3/envs/cloudspace/lib/python3.12/site-packages/transformers/modeling_attn_mask_utils.py:281: FutureWarning: The attention mask API under `transformers.modeling_attn_mask_utils` (`AttentionMaskConverter`) is deprecated and will be removed in Transformers v5.10. Please use the new API in `transformers.masking_utils`.
|
| 60 |
+
warnings.warn(DEPRECATION_MESSAGE, FutureWarning)
|
| 61 |
+
/home/zeus/miniconda3/envs/cloudspace/lib/python3.12/site-packages/transformers/modeling_attn_mask_utils.py:172: FutureWarning: The attention mask API under `transformers.modeling_attn_mask_utils` (`AttentionMaskConverter`) is deprecated and will be removed in Transformers v5.10. Please use the new API in `transformers.masking_utils`.
|
| 62 |
+
warnings.warn(DEPRECATION_MESSAGE, FutureWarning)
|
| 63 |
+
/home/zeus/miniconda3/envs/cloudspace/lib/python3.12/site-packages/transformers/modeling_attn_mask_utils.py:202: FutureWarning: The attention mask API under `transformers.modeling_attn_mask_utils` (`AttentionMaskConverter`) is deprecated and will be removed in Transformers v5.10. Please use the new API in `transformers.masking_utils`.
|
| 64 |
+
warnings.warn(DEPRECATION_MESSAGE, FutureWarning)
|
| 65 |
+
/home/zeus/miniconda3/envs/cloudspace/lib/python3.12/site-packages/transformers/modeling_attn_mask_utils.py:254: FutureWarning: The attention mask API under `transformers.modeling_attn_mask_utils` (`AttentionMaskConverter`) is deprecated and will be removed in Transformers v5.10. Please use the new API in `transformers.masking_utils`.
|
| 66 |
+
warnings.warn(DEPRECATION_MESSAGE, FutureWarning)
|
| 67 |
+
|
| 68 |
1%|β | 2/150 [01:08<1:21:40, 33.11s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 69 |
+
/home/zeus/miniconda3/envs/cloudspace/lib/python3.12/site-packages/transformers/modeling_attn_mask_utils.py:71: FutureWarning: The attention mask API under `transformers.modeling_attn_mask_utils` (`AttentionMaskConverter`) is deprecated and will be removed in Transformers v5.10. Please use the new API in `transformers.masking_utils`.
|
| 70 |
+
warnings.warn(DEPRECATION_MESSAGE, FutureWarning)
|
| 71 |
+
/home/zeus/miniconda3/envs/cloudspace/lib/python3.12/site-packages/transformers/modeling_attn_mask_utils.py:281: FutureWarning: The attention mask API under `transformers.modeling_attn_mask_utils` (`AttentionMaskConverter`) is deprecated and will be removed in Transformers v5.10. Please use the new API in `transformers.masking_utils`.
|
| 72 |
+
warnings.warn(DEPRECATION_MESSAGE, FutureWarning)
|
| 73 |
+
/home/zeus/miniconda3/envs/cloudspace/lib/python3.12/site-packages/transformers/modeling_attn_mask_utils.py:172: FutureWarning: The attention mask API under `transformers.modeling_attn_mask_utils` (`AttentionMaskConverter`) is deprecated and will be removed in Transformers v5.10. Please use the new API in `transformers.masking_utils`.
|
| 74 |
+
warnings.warn(DEPRECATION_MESSAGE, FutureWarning)
|
| 75 |
+
/home/zeus/miniconda3/envs/cloudspace/lib/python3.12/site-packages/transformers/modeling_attn_mask_utils.py:202: FutureWarning: The attention mask API under `transformers.modeling_attn_mask_utils` (`AttentionMaskConverter`) is deprecated and will be removed in Transformers v5.10. Please use the new API in `transformers.masking_utils`.
|
| 76 |
+
warnings.warn(DEPRECATION_MESSAGE, FutureWarning)
|
| 77 |
+
/home/zeus/miniconda3/envs/cloudspace/lib/python3.12/site-packages/transformers/modeling_attn_mask_utils.py:254: FutureWarning: The attention mask API under `transformers.modeling_attn_mask_utils` (`AttentionMaskConverter`) is deprecated and will be removed in Transformers v5.10. Please use the new API in `transformers.masking_utils`.
|
| 78 |
+
warnings.warn(DEPRECATION_MESSAGE, FutureWarning)
|
| 79 |
+
|
| 80 |
2%|β | 3/150 [01:55<1:36:41, 39.46s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 81 |
+
|
| 82 |
3%|β | 4/150 [02:42<1:42:40, 42.20s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 83 |
+
|
| 84 |
3%|β | 5/150 [03:00<1:20:50, 33.45s/it]
|
| 85 |
|
| 86 |
+
|
| 87 |
3%|β | 5/150 [03:00<1:20:50, 33.45s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 88 |
+
|
| 89 |
4%|β | 6/150 [03:30<1:17:51, 32.44s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 90 |
+
|
| 91 |
5%|β | 7/150 [04:02<1:16:34, 32.13s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 92 |
+
|
| 93 |
5%|β | 8/150 [04:51<1:29:21, 37.76s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 94 |
+
|
| 95 |
6%|β | 9/150 [05:40<1:36:46, 41.18s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 96 |
+
|
| 97 |
7%|β | 10/150 [06:29<1:41:47, 43.63s/it]
|
| 98 |
|
| 99 |
+
|
| 100 |
7%|β | 10/150 [06:29<1:41:47, 43.63s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 101 |
+
|
| 102 |
7%|β | 11/150 [07:00<1:31:51, 39.65s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 103 |
+
|
| 104 |
8%|β | 12/150 [07:28<1:23:12, 36.18s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 105 |
+
|
| 106 |
9%|β | 13/150 [08:17<1:31:16, 39.98s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 107 |
+
|
| 108 |
9%|β | 14/150 [08:38<1:17:25, 34.16s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 109 |
+
|
| 110 |
10%|β | 15/150 [09:27<1:27:10, 38.74s/it]
|
| 111 |
|
| 112 |
+
|
| 113 |
10%|β | 15/150 [09:27<1:27:10, 38.74s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 114 |
+
|
| 115 |
11%|β | 16/150 [10:02<1:23:47, 37.52s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 116 |
+
|
| 117 |
11%|ββ | 17/150 [10:33<1:19:19, 35.79s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 118 |
+
|
| 119 |
12%|ββ | 18/150 [11:02<1:14:05, 33.68s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 120 |
+
|
| 121 |
13%|ββ | 19/150 [11:51<1:23:25, 38.21s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 122 |
+
|
| 123 |
13%|ββ | 20/150 [12:17<1:14:37, 34.44s/it]
|
| 124 |
|
| 125 |
+
|
| 126 |
13%|ββ | 20/150 [12:17<1:14:37, 34.44s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 127 |
+
|
| 128 |
14%|ββ | 21/150 [13:05<1:23:13, 38.71s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 129 |
+
|
| 130 |
15%|ββ | 22/150 [13:54<1:29:01, 41.73s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 131 |
+
|
| 132 |
15%|ββ | 23/150 [14:19<1:17:39, 36.69s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 133 |
+
|
| 134 |
16%|ββ | 24/150 [15:07<1:24:19, 40.15s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 135 |
+
|
| 136 |
17%|ββ | 25/150 [15:27<1:11:10, 34.16s/it]
|
| 137 |
|
| 138 |
+
|
| 139 |
17%|ββ | 25/150 [15:27<1:11:10, 34.16s/it]Unsloth: Restored added_tokens_decoder metadata in outputs/grpo_adapter/checkpoint-25/tokenizer_config.json.
|
| 140 |
+
Unsloth: Restored added_tokens_decoder metadata in outputs/grpo_adapter/tokenizer_config.json.
|
| 141 |
+
Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 142 |
+
/home/zeus/miniconda3/envs/cloudspace/lib/python3.12/site-packages/transformers/modeling_attn_mask_utils.py:71: FutureWarning: The attention mask API under `transformers.modeling_attn_mask_utils` (`AttentionMaskConverter`) is deprecated and will be removed in Transformers v5.10. Please use the new API in `transformers.masking_utils`.
|
| 143 |
+
warnings.warn(DEPRECATION_MESSAGE, FutureWarning)
|
| 144 |
+
/home/zeus/miniconda3/envs/cloudspace/lib/python3.12/site-packages/transformers/modeling_attn_mask_utils.py:281: FutureWarning: The attention mask API under `transformers.modeling_attn_mask_utils` (`AttentionMaskConverter`) is deprecated and will be removed in Transformers v5.10. Please use the new API in `transformers.masking_utils`.
|
| 145 |
+
warnings.warn(DEPRECATION_MESSAGE, FutureWarning)
|
| 146 |
+
/home/zeus/miniconda3/envs/cloudspace/lib/python3.12/site-packages/transformers/modeling_attn_mask_utils.py:172: FutureWarning: The attention mask API under `transformers.modeling_attn_mask_utils` (`AttentionMaskConverter`) is deprecated and will be removed in Transformers v5.10. Please use the new API in `transformers.masking_utils`.
|
| 147 |
+
warnings.warn(DEPRECATION_MESSAGE, FutureWarning)
|
| 148 |
+
/home/zeus/miniconda3/envs/cloudspace/lib/python3.12/site-packages/transformers/modeling_attn_mask_utils.py:202: FutureWarning: The attention mask API under `transformers.modeling_attn_mask_utils` (`AttentionMaskConverter`) is deprecated and will be removed in Transformers v5.10. Please use the new API in `transformers.masking_utils`.
|
| 149 |
+
warnings.warn(DEPRECATION_MESSAGE, FutureWarning)
|
| 150 |
+
/home/zeus/miniconda3/envs/cloudspace/lib/python3.12/site-packages/transformers/modeling_attn_mask_utils.py:254: FutureWarning: The attention mask API under `transformers.modeling_attn_mask_utils` (`AttentionMaskConverter`) is deprecated and will be removed in Transformers v5.10. Please use the new API in `transformers.masking_utils`.
|
| 151 |
+
warnings.warn(DEPRECATION_MESSAGE, FutureWarning)
|
| 152 |
+
|
| 153 |
17%|ββ | 26/150 [16:05<1:13:02, 35.34s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 154 |
+
|
| 155 |
18%|ββ | 27/150 [16:37<1:10:06, 34.20s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 156 |
+
|
| 157 |
19%|ββ | 28/150 [17:27<1:19:02, 38.87s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 158 |
+
|
| 159 |
19%|ββ | 29/150 [17:56<1:12:30, 35.95s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 160 |
+
|
| 161 |
20%|ββ | 30/150 [18:24<1:07:18, 33.65s/it]
|
| 162 |
|
| 163 |
+
|
| 164 |
20%|ββ | 30/150 [18:24<1:07:18, 33.65s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 165 |
+
|
| 166 |
21%|ββ | 31/150 [18:56<1:05:26, 32.99s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 167 |
+
|
| 168 |
21%|βββ | 32/150 [19:28<1:04:17, 32.69s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 169 |
+
|
| 170 |
22%|βββ | 33/150 [19:44<53:58, 27.68s/it] Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 171 |
+
|
| 172 |
23%|βββ | 34/150 [20:24<1:00:42, 31.40s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 173 |
+
|
| 174 |
23%|βββ | 35/150 [21:14<1:11:16, 37.19s/it]
|
| 175 |
|
| 176 |
+
|
| 177 |
23%|βββ | 35/150 [21:14<1:11:16, 37.19s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 178 |
+
|
| 179 |
24%|βββ | 36/150 [21:49<1:09:00, 36.32s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 180 |
+
|
| 181 |
25%|βββ | 37/150 [22:39<1:16:10, 40.44s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 182 |
+
|
| 183 |
25%|βββ | 38/150 [23:04<1:07:08, 35.97s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 184 |
+
|
| 185 |
26%|βββ | 39/150 [23:42<1:07:27, 36.46s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 186 |
+
|
| 187 |
27%|βββ | 40/150 [24:33<1:15:08, 40.98s/it]
|
| 188 |
|
| 189 |
+
|
| 190 |
27%|βββ | 40/150 [24:33<1:15:08, 40.98s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 191 |
+
|
| 192 |
27%|βββ | 41/150 [25:00<1:06:32, 36.63s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 193 |
+
|
| 194 |
28%|βββ | 42/150 [25:18<56:04, 31.16s/it] Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 195 |
+
|
| 196 |
29%|βββ | 43/150 [25:45<53:06, 29.78s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 197 |
+
|
| 198 |
29%|βββ | 44/150 [26:19<55:08, 31.21s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 199 |
+
|
| 200 |
30%|βββ | 45/150 [27:08<1:03:36, 36.35s/it]
|
| 201 |
|
| 202 |
+
|
| 203 |
30%|βββ | 45/150 [27:08<1:03:36, 36.35s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 204 |
+
|
| 205 |
31%|βββ | 46/150 [27:57<1:09:50, 40.30s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 206 |
+
|
| 207 |
31%|ββββ | 47/150 [28:39<1:10:07, 40.85s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 208 |
+
|
| 209 |
32%|ββββ | 48/150 [29:28<1:13:14, 43.09s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 210 |
+
|
| 211 |
33%|ββββ | 49/150 [30:16<1:15:10, 44.66s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 212 |
+
|
| 213 |
33%|ββββ | 50/150 [30:42<1:04:53, 38.93s/it]
|
| 214 |
|
| 215 |
+
|
| 216 |
33%|ββββ | 50/150 [30:42<1:04:53, 38.93s/it]Unsloth: Restored added_tokens_decoder metadata in outputs/grpo_adapter/checkpoint-50/tokenizer_config.json.
|
| 217 |
+
Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 218 |
+
/home/zeus/miniconda3/envs/cloudspace/lib/python3.12/site-packages/transformers/modeling_attn_mask_utils.py:71: FutureWarning: The attention mask API under `transformers.modeling_attn_mask_utils` (`AttentionMaskConverter`) is deprecated and will be removed in Transformers v5.10. Please use the new API in `transformers.masking_utils`.
|
| 219 |
+
warnings.warn(DEPRECATION_MESSAGE, FutureWarning)
|
| 220 |
+
/home/zeus/miniconda3/envs/cloudspace/lib/python3.12/site-packages/transformers/modeling_attn_mask_utils.py:281: FutureWarning: The attention mask API under `transformers.modeling_attn_mask_utils` (`AttentionMaskConverter`) is deprecated and will be removed in Transformers v5.10. Please use the new API in `transformers.masking_utils`.
|
| 221 |
+
warnings.warn(DEPRECATION_MESSAGE, FutureWarning)
|
| 222 |
+
/home/zeus/miniconda3/envs/cloudspace/lib/python3.12/site-packages/transformers/modeling_attn_mask_utils.py:172: FutureWarning: The attention mask API under `transformers.modeling_attn_mask_utils` (`AttentionMaskConverter`) is deprecated and will be removed in Transformers v5.10. Please use the new API in `transformers.masking_utils`.
|
| 223 |
+
warnings.warn(DEPRECATION_MESSAGE, FutureWarning)
|
| 224 |
+
/home/zeus/miniconda3/envs/cloudspace/lib/python3.12/site-packages/transformers/modeling_attn_mask_utils.py:202: FutureWarning: The attention mask API under `transformers.modeling_attn_mask_utils` (`AttentionMaskConverter`) is deprecated and will be removed in Transformers v5.10. Please use the new API in `transformers.masking_utils`.
|
| 225 |
+
warnings.warn(DEPRECATION_MESSAGE, FutureWarning)
|
| 226 |
+
/home/zeus/miniconda3/envs/cloudspace/lib/python3.12/site-packages/transformers/modeling_attn_mask_utils.py:254: FutureWarning: The attention mask API under `transformers.modeling_attn_mask_utils` (`AttentionMaskConverter`) is deprecated and will be removed in Transformers v5.10. Please use the new API in `transformers.masking_utils`.
|
| 227 |
+
warnings.warn(DEPRECATION_MESSAGE, FutureWarning)
|
| 228 |
+
|
| 229 |
34%|ββββ | 51/150 [31:11<59:43, 36.20s/it] Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 230 |
+
|
| 231 |
35%|ββββ | 52/150 [31:40<55:17, 33.85s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 232 |
+
|
| 233 |
35%|ββββ | 53/150 [32:05<50:19, 31.13s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 234 |
+
|
| 235 |
36%|ββββ | 54/150 [32:54<58:45, 36.72s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 236 |
+
|
| 237 |
37%|ββββ | 55/150 [33:43<1:03:55, 40.38s/it]
|
| 238 |
|
| 239 |
+
|
| 240 |
37%|ββββ | 55/150 [33:43<1:03:55, 40.38s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 241 |
+
|
| 242 |
37%|ββββ | 56/150 [34:14<58:58, 37.64s/it] Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 243 |
+
|
| 244 |
38%|ββββ | 57/150 [34:38<51:44, 33.38s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 245 |
+
|
| 246 |
39%|ββββ | 58/150 [35:27<58:16, 38.01s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 247 |
+
|
| 248 |
39%|ββββ | 59/150 [35:58<54:42, 36.07s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 249 |
+
|
| 250 |
40%|ββββ | 60/150 [36:30<51:58, 34.65s/it]
|
| 251 |
|
| 252 |
+
|
| 253 |
40%|ββββ | 60/150 [36:30<51:58, 34.65s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 254 |
+
|
| 255 |
41%|ββββ | 61/150 [37:19<57:47, 38.96s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 256 |
+
|
| 257 |
41%|βββββ | 62/150 [38:08<1:01:50, 42.17s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 258 |
+
|
| 259 |
42%|βββββ | 63/150 [38:31<52:35, 36.27s/it] Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 260 |
+
|
| 261 |
43%|βββββ | 64/150 [38:59<48:25, 33.79s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 262 |
+
|
| 263 |
43%|βββββ | 65/150 [39:21<43:05, 30.42s/it]
|
| 264 |
|
| 265 |
+
|
| 266 |
43%|βββββ | 65/150 [39:21<43:05, 30.42s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 267 |
+
|
| 268 |
44%|βββββ | 66/150 [40:10<50:23, 36.00s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 269 |
+
|
| 270 |
45%|βββββ | 67/150 [40:59<55:14, 39.93s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 271 |
+
|
| 272 |
45%|βββββ | 68/150 [41:31<51:02, 37.35s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 273 |
+
|
| 274 |
46%|βββββ | 69/150 [42:19<54:59, 40.74s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 275 |
+
|
| 276 |
47%|βββββ | 70/150 [43:03<55:21, 41.52s/it]
|
| 277 |
|
| 278 |
+
|
| 279 |
47%|βββββ | 70/150 [43:03<55:21, 41.52s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 280 |
+
|
| 281 |
47%|βββββ | 71/150 [43:34<50:33, 38.40s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 282 |
+
|
| 283 |
48%|βββββ | 72/150 [44:23<54:00, 41.55s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 284 |
+
|
| 285 |
49%|βββββ | 73/150 [44:48<46:52, 36.53s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 286 |
+
|
| 287 |
49%|βββββ | 74/150 [45:26<46:53, 37.02s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 288 |
+
|
| 289 |
50%|βββββ | 75/150 [45:48<40:55, 32.74s/it]
|
| 290 |
|
| 291 |
+
|
| 292 |
50%|βββββ | 75/150 [45:48<40:55, 32.74s/it]Unsloth: Restored added_tokens_decoder metadata in outputs/grpo_adapter/checkpoint-75/tokenizer_config.json.
|
| 293 |
+
Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 294 |
+
/home/zeus/miniconda3/envs/cloudspace/lib/python3.12/site-packages/transformers/modeling_attn_mask_utils.py:71: FutureWarning: The attention mask API under `transformers.modeling_attn_mask_utils` (`AttentionMaskConverter`) is deprecated and will be removed in Transformers v5.10. Please use the new API in `transformers.masking_utils`.
|
| 295 |
+
warnings.warn(DEPRECATION_MESSAGE, FutureWarning)
|
| 296 |
+
/home/zeus/miniconda3/envs/cloudspace/lib/python3.12/site-packages/transformers/modeling_attn_mask_utils.py:281: FutureWarning: The attention mask API under `transformers.modeling_attn_mask_utils` (`AttentionMaskConverter`) is deprecated and will be removed in Transformers v5.10. Please use the new API in `transformers.masking_utils`.
|
| 297 |
+
warnings.warn(DEPRECATION_MESSAGE, FutureWarning)
|
| 298 |
+
/home/zeus/miniconda3/envs/cloudspace/lib/python3.12/site-packages/transformers/modeling_attn_mask_utils.py:172: FutureWarning: The attention mask API under `transformers.modeling_attn_mask_utils` (`AttentionMaskConverter`) is deprecated and will be removed in Transformers v5.10. Please use the new API in `transformers.masking_utils`.
|
| 299 |
+
warnings.warn(DEPRECATION_MESSAGE, FutureWarning)
|
| 300 |
+
/home/zeus/miniconda3/envs/cloudspace/lib/python3.12/site-packages/transformers/modeling_attn_mask_utils.py:202: FutureWarning: The attention mask API under `transformers.modeling_attn_mask_utils` (`AttentionMaskConverter`) is deprecated and will be removed in Transformers v5.10. Please use the new API in `transformers.masking_utils`.
|
| 301 |
+
warnings.warn(DEPRECATION_MESSAGE, FutureWarning)
|
| 302 |
+
/home/zeus/miniconda3/envs/cloudspace/lib/python3.12/site-packages/transformers/modeling_attn_mask_utils.py:254: FutureWarning: The attention mask API under `transformers.modeling_attn_mask_utils` (`AttentionMaskConverter`) is deprecated and will be removed in Transformers v5.10. Please use the new API in `transformers.masking_utils`.
|
| 303 |
+
warnings.warn(DEPRECATION_MESSAGE, FutureWarning)
|
| 304 |
+
|
| 305 |
51%|βββββ | 76/150 [46:14<37:39, 30.53s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 306 |
+
|
| 307 |
51%|ββββββ | 77/150 [47:03<43:48, 36.01s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 308 |
+
|
| 309 |
52%|ββββββ | 78/150 [47:38<42:48, 35.67s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 310 |
+
|
| 311 |
53%|ββββββ | 79/150 [48:29<47:49, 40.41s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 312 |
+
|
| 313 |
53%|ββββββ | 80/150 [49:17<49:54, 42.78s/it]
|
| 314 |
|
| 315 |
+
|
| 316 |
53%|ββββββ | 80/150 [49:17<49:54, 42.78s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 317 |
+
|
| 318 |
54%|ββββββ | 81/150 [49:49<45:15, 39.36s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 319 |
+
|
| 320 |
55%|ββββββ | 82/150 [50:15<40:14, 35.50s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 321 |
+
|
| 322 |
55%|ββββββ | 83/150 [50:48<38:35, 34.56s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 323 |
+
|
| 324 |
56%|ββββββ | 84/150 [51:36<42:34, 38.70s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 325 |
+
|
| 326 |
57%|ββββββ | 85/150 [52:07<39:26, 36.41s/it]
|
| 327 |
|
| 328 |
+
|
| 329 |
57%|ββββββ | 85/150 [52:07<39:26, 36.41s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 330 |
+
|
| 331 |
57%|ββββββ | 86/150 [52:33<35:23, 33.18s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 332 |
+
|
| 333 |
58%|ββββββ | 87/150 [53:07<35:14, 33.56s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 334 |
+
|
| 335 |
59%|ββββββ | 88/150 [53:29<31:01, 30.03s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 336 |
+
|
| 337 |
59%|ββββββ | 89/150 [54:10<33:57, 33.39s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 338 |
+
|
| 339 |
60%|ββββββ | 90/150 [54:38<31:37, 31.63s/it]
|
| 340 |
|
| 341 |
+
|
| 342 |
60%|ββββββ | 90/150 [54:38<31:37, 31.63s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 343 |
+
|
| 344 |
61%|ββββββ | 91/150 [55:06<30:02, 30.55s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 345 |
+
|
| 346 |
61%|βββββββ | 92/150 [55:33<28:28, 29.46s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 347 |
+
|
| 348 |
62%|βββββββ | 93/150 [56:23<33:49, 35.61s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 349 |
+
|
| 350 |
63%|βββββββ | 94/150 [56:53<31:55, 34.21s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 351 |
+
|
| 352 |
63%|βββββββ | 95/150 [57:19<29:00, 31.64s/it]
|
| 353 |
|
| 354 |
+
|
| 355 |
63%|βββββββ | 95/150 [57:19<29:00, 31.64s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 356 |
+
|
| 357 |
64%|βββββββ | 96/150 [57:55<29:34, 32.87s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 358 |
+
|
| 359 |
65%|βββββββ | 97/150 [58:43<32:58, 37.33s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 360 |
+
|
| 361 |
65%|βββββββ | 98/150 [59:17<31:36, 36.48s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 362 |
+
|
| 363 |
66%|βββββββ | 99/150 [1:00:06<34:10, 40.20s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 364 |
+
|
| 365 |
67%|βββββββ | 100/150 [1:00:48<34:02, 40.85s/it]
|
| 366 |
|
| 367 |
+
|
| 368 |
67%|βββββββ | 100/150 [1:00:48<34:02, 40.85s/it]Unsloth: Restored added_tokens_decoder metadata in outputs/grpo_adapter/checkpoint-100/tokenizer_config.json.
|
| 369 |
+
Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 370 |
+
/home/zeus/miniconda3/envs/cloudspace/lib/python3.12/site-packages/transformers/modeling_attn_mask_utils.py:71: FutureWarning: The attention mask API under `transformers.modeling_attn_mask_utils` (`AttentionMaskConverter`) is deprecated and will be removed in Transformers v5.10. Please use the new API in `transformers.masking_utils`.
|
| 371 |
+
warnings.warn(DEPRECATION_MESSAGE, FutureWarning)
|
| 372 |
+
/home/zeus/miniconda3/envs/cloudspace/lib/python3.12/site-packages/transformers/modeling_attn_mask_utils.py:281: FutureWarning: The attention mask API under `transformers.modeling_attn_mask_utils` (`AttentionMaskConverter`) is deprecated and will be removed in Transformers v5.10. Please use the new API in `transformers.masking_utils`.
|
| 373 |
+
warnings.warn(DEPRECATION_MESSAGE, FutureWarning)
|
| 374 |
+
/home/zeus/miniconda3/envs/cloudspace/lib/python3.12/site-packages/transformers/modeling_attn_mask_utils.py:172: FutureWarning: The attention mask API under `transformers.modeling_attn_mask_utils` (`AttentionMaskConverter`) is deprecated and will be removed in Transformers v5.10. Please use the new API in `transformers.masking_utils`.
|
| 375 |
+
warnings.warn(DEPRECATION_MESSAGE, FutureWarning)
|
| 376 |
+
/home/zeus/miniconda3/envs/cloudspace/lib/python3.12/site-packages/transformers/modeling_attn_mask_utils.py:202: FutureWarning: The attention mask API under `transformers.modeling_attn_mask_utils` (`AttentionMaskConverter`) is deprecated and will be removed in Transformers v5.10. Please use the new API in `transformers.masking_utils`.
|
| 377 |
+
warnings.warn(DEPRECATION_MESSAGE, FutureWarning)
|
| 378 |
+
/home/zeus/miniconda3/envs/cloudspace/lib/python3.12/site-packages/transformers/modeling_attn_mask_utils.py:254: FutureWarning: The attention mask API under `transformers.modeling_attn_mask_utils` (`AttentionMaskConverter`) is deprecated and will be removed in Transformers v5.10. Please use the new API in `transformers.masking_utils`.
|
| 379 |
+
warnings.warn(DEPRECATION_MESSAGE, FutureWarning)
|
| 380 |
+
|
| 381 |
67%|βββββββ | 101/150 [1:01:12<29:02, 35.56s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 382 |
+
|
| 383 |
68%|βββββββ | 102/150 [1:02:00<31:34, 39.47s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 384 |
+
|
| 385 |
69%|βββββββ | 103/150 [1:02:28<28:07, 35.91s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 386 |
+
|
| 387 |
69%|βββββββ | 104/150 [1:03:02<27:08, 35.41s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 388 |
+
|
| 389 |
70%|βββββββ | 105/150 [1:03:51<29:34, 39.43s/it]
|
| 390 |
|
| 391 |
+
|
| 392 |
70%|βββββββ | 105/150 [1:03:51<29:34, 39.43s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 393 |
+
|
| 394 |
71%|βββββββ | 106/150 [1:04:39<30:55, 42.18s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 395 |
+
|
| 396 |
71%|ββββββββ | 107/150 [1:05:27<31:29, 43.93s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 397 |
+
|
| 398 |
72%|ββββββββ | 108/150 [1:05:47<25:35, 36.57s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 399 |
+
|
| 400 |
73%|ββββββββ | 109/150 [1:06:32<26:51, 39.30s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 401 |
+
|
| 402 |
73%|ββββββββ | 110/150 [1:07:01<23:56, 35.92s/it]
|
| 403 |
|
| 404 |
+
|
| 405 |
73%|ββββββββ | 110/150 [1:07:01<23:56, 35.92s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 406 |
+
|
| 407 |
74%|ββββββββ | 111/150 [1:07:27<21:36, 33.23s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 408 |
+
|
| 409 |
75%|ββββββββ | 112/150 [1:07:53<19:29, 30.77s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 410 |
+
|
| 411 |
75%|ββββββββ | 113/150 [1:08:23<19:00, 30.82s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 412 |
+
|
| 413 |
76%|ββββββββ | 114/150 [1:09:12<21:40, 36.13s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 414 |
+
|
| 415 |
77%|ββββββββ | 115/150 [1:09:51<21:36, 37.05s/it]
|
| 416 |
|
| 417 |
+
|
| 418 |
77%|ββββββββ | 115/150 [1:09:51<21:36, 37.05s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 419 |
+
|
| 420 |
77%|ββββββββ | 116/150 [1:10:22<20:01, 35.34s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 421 |
+
|
| 422 |
78%|ββββββββ | 117/150 [1:11:19<22:54, 41.66s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 423 |
+
|
| 424 |
79%|ββββββββ | 118/150 [1:11:39<18:44, 35.16s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 425 |
+
|
| 426 |
79%|ββββββββ | 119/150 [1:12:27<20:09, 39.03s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 427 |
+
|
| 428 |
80%|ββββββββ | 120/150 [1:12:56<17:56, 35.89s/it]
|
| 429 |
|
| 430 |
+
|
| 431 |
80%|ββββββββ | 120/150 [1:12:56<17:56, 35.89s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 432 |
+
|
| 433 |
81%|ββββββββ | 121/150 [1:13:53<20:28, 42.37s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 434 |
+
|
| 435 |
81%|βββββββββ | 122/150 [1:14:42<20:40, 44.31s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 436 |
+
|
| 437 |
82%|βββββββββ | 123/150 [1:15:22<19:21, 43.01s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 438 |
+
|
| 439 |
83%|βββββββββ | 124/150 [1:15:57<17:36, 40.62s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 440 |
+
|
| 441 |
83%|βββββββββ | 125/150 [1:16:46<17:56, 43.06s/it]
|
| 442 |
|
| 443 |
+
|
| 444 |
83%|βββββββββ | 125/150 [1:16:46<17:56, 43.06s/it]Unsloth: Restored added_tokens_decoder metadata in outputs/grpo_adapter/checkpoint-125/tokenizer_config.json.
|
| 445 |
+
Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 446 |
+
/home/zeus/miniconda3/envs/cloudspace/lib/python3.12/site-packages/transformers/modeling_attn_mask_utils.py:71: FutureWarning: The attention mask API under `transformers.modeling_attn_mask_utils` (`AttentionMaskConverter`) is deprecated and will be removed in Transformers v5.10. Please use the new API in `transformers.masking_utils`.
|
| 447 |
+
warnings.warn(DEPRECATION_MESSAGE, FutureWarning)
|
| 448 |
+
/home/zeus/miniconda3/envs/cloudspace/lib/python3.12/site-packages/transformers/modeling_attn_mask_utils.py:281: FutureWarning: The attention mask API under `transformers.modeling_attn_mask_utils` (`AttentionMaskConverter`) is deprecated and will be removed in Transformers v5.10. Please use the new API in `transformers.masking_utils`.
|
| 449 |
+
warnings.warn(DEPRECATION_MESSAGE, FutureWarning)
|
| 450 |
+
/home/zeus/miniconda3/envs/cloudspace/lib/python3.12/site-packages/transformers/modeling_attn_mask_utils.py:172: FutureWarning: The attention mask API under `transformers.modeling_attn_mask_utils` (`AttentionMaskConverter`) is deprecated and will be removed in Transformers v5.10. Please use the new API in `transformers.masking_utils`.
|
| 451 |
+
warnings.warn(DEPRECATION_MESSAGE, FutureWarning)
|
| 452 |
+
/home/zeus/miniconda3/envs/cloudspace/lib/python3.12/site-packages/transformers/modeling_attn_mask_utils.py:202: FutureWarning: The attention mask API under `transformers.modeling_attn_mask_utils` (`AttentionMaskConverter`) is deprecated and will be removed in Transformers v5.10. Please use the new API in `transformers.masking_utils`.
|
| 453 |
+
warnings.warn(DEPRECATION_MESSAGE, FutureWarning)
|
| 454 |
+
/home/zeus/miniconda3/envs/cloudspace/lib/python3.12/site-packages/transformers/modeling_attn_mask_utils.py:254: FutureWarning: The attention mask API under `transformers.modeling_attn_mask_utils` (`AttentionMaskConverter`) is deprecated and will be removed in Transformers v5.10. Please use the new API in `transformers.masking_utils`.
|
| 455 |
+
warnings.warn(DEPRECATION_MESSAGE, FutureWarning)
|
| 456 |
+
|
| 457 |
84%|βββββββββ | 126/150 [1:17:36<18:04, 45.19s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 458 |
+
|
| 459 |
85%|βββββββββ | 127/150 [1:18:10<16:02, 41.86s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 460 |
+
|
| 461 |
85%|βββββββββ | 128/150 [1:18:36<13:39, 37.24s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 462 |
+
|
| 463 |
86%|βββββββββ | 129/150 [1:19:26<14:22, 41.05s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 464 |
+
|
| 465 |
87%|βββββββββ | 130/150 [1:19:48<11:44, 35.24s/it]
|
| 466 |
|
| 467 |
+
|
| 468 |
87%|βββββββββ | 130/150 [1:19:48<11:44, 35.24s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 469 |
+
|
| 470 |
87%|βββββββββ | 131/150 [1:20:37<12:26, 39.31s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 471 |
+
|
| 472 |
88%|βββββββββ | 132/150 [1:21:08<11:03, 36.86s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 473 |
+
|
| 474 |
89%|βββββββββ | 133/150 [1:21:34<09:31, 33.63s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 475 |
+
|
| 476 |
89%|βββββββββ | 134/150 [1:21:59<08:15, 30.99s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 477 |
+
|
| 478 |
90%|βββββββββ | 135/150 [1:22:35<08:07, 32.50s/it]
|
| 479 |
|
| 480 |
+
|
| 481 |
90%|βββββββββ | 135/150 [1:22:35<08:07, 32.50s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 482 |
+
|
| 483 |
91%|βββββββββ | 136/150 [1:23:07<07:34, 32.49s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 484 |
+
|
| 485 |
91%|ββββββββββ| 137/150 [1:23:56<08:05, 37.35s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 486 |
+
|
| 487 |
92%|ββββββββββ| 138/150 [1:24:25<06:56, 34.73s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 488 |
+
|
| 489 |
93%|ββββββββββ| 139/150 [1:25:05<06:40, 36.39s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 490 |
+
|
| 491 |
93%|ββββββββββ| 140/150 [1:25:40<05:59, 35.92s/it]
|
| 492 |
|
| 493 |
+
|
| 494 |
93%|ββββββββββ| 140/150 [1:25:40<05:59, 35.92s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 495 |
+
|
| 496 |
94%|ββββββββββ| 141/150 [1:26:00<04:41, 31.23s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 497 |
+
|
| 498 |
95%|ββββββββββ| 142/150 [1:26:24<03:51, 28.91s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 499 |
+
|
| 500 |
95%|ββββββββββ| 143/150 [1:26:43<03:01, 25.95s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 501 |
+
|
| 502 |
96%|ββββββββββ| 144/150 [1:27:16<02:49, 28.33s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 503 |
+
|
| 504 |
97%|ββββββββββ| 145/150 [1:28:05<02:51, 34.29s/it]
|
| 505 |
|
| 506 |
+
|
| 507 |
97%|ββββββββββ| 145/150 [1:28:05<02:51, 34.29s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 508 |
+
|
| 509 |
97%|ββββββββββ| 146/150 [1:28:35<02:12, 33.15s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 510 |
+
|
| 511 |
98%|ββββββββββ| 147/150 [1:29:07<01:38, 32.68s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 512 |
+
|
| 513 |
99%|ββββββββββ| 148/150 [1:29:32<01:00, 30.38s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 514 |
+
|
| 515 |
99%|ββββββββββ| 149/150 [1:30:21<00:35, 35.90s/it]Both `max_new_tokens` (=1024) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 516 |
+
|
| 517 |
|
| 518 |
+
|
| 519 |
+
|
| 520 |
|
| 521 |
+
|
| 522 |
+
|
| 523 |
+
|
| 524 |
+
|
| 525 |
+
|
| 526 |
...adapter/training_args.bin: 100%|ββββββββββ| 6.74kB / 6.74kB [A[A
|
| 527 |
+
|
| 528 |
+
|
| 529 |
+
|
| 530 |
...po_adapter/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB [A[A[A
|
| 531 |
+
|
| 532 |
+
|
| 533 |
+
|
| 534 |
+
|
| 535 |
...adapter_model.safetensors: 100%|ββββββββββ| 162MB / 162MB [A[A[A[A
|
| 536 |
+
|
| 537 |
+
|
| 538 |
...adapter/training_args.bin: 100%|ββββββββββ| 6.74kB / 6.74kB [A[A
|
| 539 |
+
|
| 540 |
+
|
| 541 |
+
|
| 542 |
...po_adapter/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB [A[A[A
|
| 543 |
+
|
| 544 |
+
|
| 545 |
+
|
| 546 |
+
|
| 547 |
...adapter_model.safetensors: 100%|ββββββββββ| 162MB / 162MB [A[A[A[A
|
| 548 |
+
|
| 549 |
+
|
| 550 |
...adapter/training_args.bin: 100%|ββββββββββ| 6.74kB / 6.74kB [A[A
|
| 551 |
+
|
| 552 |
+
|
| 553 |
+
|
| 554 |
...po_adapter/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB [A[A[A
|
| 555 |
+
|
| 556 |
+
|
| 557 |
+
|
| 558 |
+
|
| 559 |
...adapter_model.safetensors: 100%|ββββββββββ| 162MB / 162MB [A[A[A[A
|
| 560 |
+
|
| 561 |
+
|
| 562 |
...adapter/training_args.bin: 100%|ββββββββββ| 6.74kB / 6.74kB [A[A
|
| 563 |
+
|
| 564 |
+
|
| 565 |
+
|
| 566 |
...po_adapter/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB [A[A[A
|
| 567 |
+
|
| 568 |
+
|
| 569 |
+
|
| 570 |
+
|
| 571 |
...adapter_model.safetensors: 100%|ββββββββββ| 162MB / 162MB [A[A[A[A
|
| 572 |
+
|
| 573 |
+
|
| 574 |
...adapter/training_args.bin: 100%|ββββββββββ| 6.74kB / 6.74kB [A[A
|
| 575 |
+
|
| 576 |
+
|
| 577 |
+
|
| 578 |
...po_adapter/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB [A[A[A
|
| 579 |
+
|
| 580 |
+
|
| 581 |
+
|
| 582 |
+
|
| 583 |
...adapter_model.safetensors: 100%|ββββββββββ| 162MB / 162MB [A[A[A[A
|
| 584 |
+
|
| 585 |
+
|
| 586 |
...adapter/training_args.bin: 100%|ββββββββββ| 6.74kB / 6.74kB
|
| 587 |
+
|
| 588 |
...po_adapter/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB
|
| 589 |
+
|
| 590 |
...adapter_model.safetensors: 100%|ββββββββββ| 162MB / 162MB
|
| 591 |
+
No files have been modified since last commit. Skipping to prevent empty commit.
|
| 592 |
+
[huggingface_hub.hf_api|WARNING]No files have been modified since last commit. Skipping to prevent empty commit.
|
| 593 |
+
|
| 594 |
+
|
| 595 |
+
|
| 596 |
+
|
| 597 |
...adapter/training_args.bin: 100%|ββββββββββ| 6.74kB / 6.74kB [A[A
|
| 598 |
+
|
| 599 |
+
|
| 600 |
+
|
| 601 |
...po_adapter/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB [A[A[A
|
| 602 |
+
|
| 603 |
+
|
| 604 |
+
|
| 605 |
+
|
| 606 |
...adapter_model.safetensors: 100%|ββββββββββ| 162MB / 162MB [A[A[A[A
|
| 607 |
+
|
| 608 |
+
|
| 609 |
...adapter/training_args.bin: 100%|ββββββββββ| 6.74kB / 6.74kB [A[A
|
| 610 |
+
|
| 611 |
+
|
| 612 |
+
|
| 613 |
...po_adapter/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB [A[A[A
|
| 614 |
+
|
| 615 |
+
|
| 616 |
+
|
| 617 |
+
|
| 618 |
...adapter_model.safetensors: 100%|ββββββββββ| 162MB / 162MB [A[A[A[A
|
| 619 |
+
|
| 620 |
+
|
| 621 |
...adapter/training_args.bin: 100%|ββββββββββ| 6.74kB / 6.74kB [A[A
|
| 622 |
+
|
| 623 |
+
|
| 624 |
+
|
| 625 |
...po_adapter/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB [A[A[A
|
| 626 |
+
|
| 627 |
+
|
| 628 |
+
|
| 629 |
+
|
| 630 |
...adapter_model.safetensors: 100%|ββββββββββ| 162MB / 162MB [A[A[A[A
|
| 631 |
+
|
| 632 |
+
|
| 633 |
...adapter/training_args.bin: 100%|ββββββββββ| 6.74kB / 6.74kB [A[A
|
| 634 |
+
|
| 635 |
+
|
| 636 |
+
|
| 637 |
...po_adapter/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB [A[A[A
|
| 638 |
+
|
| 639 |
+
|
| 640 |
+
|
| 641 |
+
|
| 642 |
...adapter_model.safetensors: 100%|ββββββββββ| 162MB / 162MB [A[A[A[A
|
| 643 |
+
|
| 644 |
+
|
| 645 |
...adapter/training_args.bin: 100%|ββββββββββ| 6.74kB / 6.74kB
|
| 646 |
+
|
| 647 |
...po_adapter/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB
|
| 648 |
+
|
| 649 |
...adapter_model.safetensors: 100%|ββββββββββ| 162MB / 162MB
|
| 650 |
+
No files have been modified since last commit. Skipping to prevent empty commit.
|
| 651 |
+
[huggingface_hub.hf_api|WARNING]No files have been modified since last commit. Skipping to prevent empty commit.
|
| 652 |
+
|
| 653 |
+
---
|
| 654 |
+
exit_code: 0
|
| 655 |
+
elapsed_ms: 5492606
|
| 656 |
+
ended_at: 2026-04-25T18:20:08.954Z
|
| 657 |
+
---
|
training_logs/221184.txt
ADDED
|
@@ -0,0 +1,48 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
pid: 85217
|
| 3 |
+
cwd: "/teamspace/studios/this_studio/corp_gym"
|
| 4 |
+
command: "python -u eval.py --policy hf --label base-deepseek-14b --model /teamspace/studios/this_studio/.cache/huggingface/hub/models--unsloth--deepseek-r1-distill-qwen-14b-unsloth-bnb-4bit/snapshots/13ae1648fa81751417bd1701a3e670c70eaa0320 --output results/runs/deepseek_14b__base/base-deepseek-14b_eval.jsonl"
|
| 5 |
+
started_at: 2026-04-26T04:12:25.855Z
|
| 6 |
+
running_for_ms: 545018
|
| 7 |
+
---
|
| 8 |
+
Skipping import of cpp extensions due to incompatible torch version. Please upgrade to torch >= 2.11.0 (found 2.7.1+cu128).
|
| 9 |
+
`torch_dtype` is deprecated! Use `dtype` instead!
|
| 10 |
+
|
| 11 |
+
The following generation flags are not valid and may be ignored: ['temperature', 'top_p']. Set `TRANSFORMERS_VERBOSITY=info` for more details.
|
| 12 |
+
Both `max_new_tokens` (=1536) and `max_length`(=131072) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 13 |
+
Both `max_new_tokens` (=1536) and `max_length`(=131072) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 14 |
+
Both `max_new_tokens` (=1536) and `max_length`(=131072) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 15 |
+
Both `max_new_tokens` (=1536) and `max_length`(=131072) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 16 |
+
Both `max_new_tokens` (=1536) and `max_length`(=131072) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 17 |
+
Both `max_new_tokens` (=1536) and `max_length`(=131072) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 18 |
+
Both `max_new_tokens` (=1536) and `max_length`(=131072) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 19 |
+
Both `max_new_tokens` (=1536) and `max_length`(=131072) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 20 |
+
Both `max_new_tokens` (=1536) and `max_length`(=131072) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 21 |
+
Both `max_new_tokens` (=1536) and `max_length`(=131072) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 22 |
+
Both `max_new_tokens` (=1536) and `max_length`(=131072) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 23 |
+
Both `max_new_tokens` (=1536) and `max_length`(=131072) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 24 |
+
Both `max_new_tokens` (=1536) and `max_length`(=131072) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 25 |
+
Both `max_new_tokens` (=1536) and `max_length`(=131072) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 26 |
+
Both `max_new_tokens` (=1536) and `max_length`(=131072) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 27 |
+
Both `max_new_tokens` (=1536) and `max_length`(=131072) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 28 |
+
Both `max_new_tokens` (=1536) and `max_length`(=131072) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 29 |
+
Both `max_new_tokens` (=1536) and `max_length`(=131072) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 30 |
+
Both `max_new_tokens` (=1536) and `max_length`(=131072) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 31 |
+
Both `max_new_tokens` (=1536) and `max_length`(=131072) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 32 |
+
Both `max_new_tokens` (=1536) and `max_length`(=131072) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 33 |
+
Both `max_new_tokens` (=1536) and `max_length`(=131072) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 34 |
+
Both `max_new_tokens` (=1536) and `max_length`(=131072) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 35 |
+
|
| 36 |
+
[base-deepseek-14b] 3 episodes
|
| 37 |
+
- e1_launch_readiness: terminal_reward=0.840 pass_rate=1.000 success=1.000
|
| 38 |
+
- m1_budget_reallocation: terminal_reward=0.807 pass_rate=0.667 success=0.000
|
| 39 |
+
- h1_acquisition_defence: terminal_reward=0.843 pass_rate=0.750 success=0.000
|
| 40 |
+
|
| 41 |
+
Wrote results/runs/deepseek_14b__base/base-deepseek-14b_eval.jsonl
|
| 42 |
+
Wrote results/runs/deepseek_14b__base/metadata.json
|
| 43 |
+
|
| 44 |
+
---
|
| 45 |
+
exit_code: 0
|
| 46 |
+
elapsed_ms: 549166
|
| 47 |
+
ended_at: 2026-04-26T04:21:35.021Z
|
| 48 |
+
---
|
training_logs/288278.txt
ADDED
|
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
pid: 116258
|
| 3 |
+
cwd: "/teamspace/studios/this_studio/corp_gym"
|
| 4 |
+
command: "stdbuf -oL -eL python -u eval.py --policy hf --label base --model Qwen/Qwen2.5-7B-Instruct --output results/base_eval.jsonl 2>&1 | tee results/runs/base_eval.log"
|
| 5 |
+
started_at: 2026-04-25T16:36:14.259Z
|
| 6 |
+
running_for_ms: 60008
|
| 7 |
+
---
|
| 8 |
+
Skipping import of cpp extensions due to incompatible torch version. Please upgrade to torch >= 2.11.0 (found 2.7.1+cu128).
|
| 9 |
+
`torch_dtype` is deprecated! Use `dtype` instead!
|
| 10 |
+
|
| 11 |
+
The following generation flags are not valid and may be ignored: ['temperature', 'top_p', 'top_k']. Set `TRANSFORMERS_VERBOSITY=info` for more details.
|
| 12 |
+
|
| 13 |
+
[base] 3 episodes
|
| 14 |
+
- e1_launch_readiness: terminal_reward=0.910 pass_rate=1.000 success=1.000
|
| 15 |
+
- m1_budget_reallocation: terminal_reward=0.807 pass_rate=0.667 success=0.000
|
| 16 |
+
- h1_acquisition_defence: terminal_reward=0.782 pass_rate=0.750 success=0.000
|
| 17 |
+
|
| 18 |
+
Wrote results/base_eval.jsonl
|
| 19 |
+
Wrote results/metadata.json
|
| 20 |
+
|
| 21 |
+
---
|
| 22 |
+
exit_code: 0
|
| 23 |
+
elapsed_ms: 64394
|
| 24 |
+
ended_at: 2026-04-25T16:37:18.653Z
|
| 25 |
+
---
|
training_logs/309711.txt
ADDED
|
@@ -0,0 +1,155 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 0 |
0%| | 0/56 [00:00<?, ?it/s]`use_return_dict` is deprecated! Use `return_dict` instead!
|
|
|
|
|
|
|
| 1 |
2%|β | 1/56 [00:11<10:36, 11.57s/it]
|
| 2 |
4%|β | 2/56 [00:23<10:39, 11.85s/it]
|
| 3 |
5%|β | 3/56 [00:38<11:40, 13.21s/it]
|
| 4 |
7%|β | 4/56 [00:51<11:27, 13.21s/it]
|
| 5 |
9%|β | 5/56 [01:07<11:53, 13.99s/it]
|
| 6 |
|
|
|
|
| 7 |
9%|β | 5/56 [01:07<11:53, 13.99s/it]
|
| 8 |
11%|β | 6/56 [01:25<12:48, 15.37s/it]
|
| 9 |
12%|ββ | 7/56 [01:40<12:31, 15.34s/it]
|
| 10 |
14%|ββ | 8/56 [01:59<13:06, 16.39s/it]
|
| 11 |
16%|ββ | 9/56 [02:18<13:40, 17.47s/it]
|
| 12 |
18%|ββ | 10/56 [02:33<12:45, 16.64s/it]
|
| 13 |
|
|
|
|
| 14 |
18%|ββ | 10/56 [02:33<12:45, 16.64s/it]
|
| 15 |
20%|ββ | 11/56 [02:52<12:56, 17.25s/it]
|
| 16 |
21%|βββ | 12/56 [03:07<12:18, 16.78s/it]
|
| 17 |
23%|βββ | 13/56 [03:18<10:43, 14.96s/it]
|
| 18 |
25%|βββ | 14/56 [03:31<10:02, 14.34s/it]
|
| 19 |
27%|βββ | 15/56 [03:46<09:49, 14.37s/it]
|
| 20 |
|
|
|
|
| 21 |
27%|βββ | 15/56 [03:46<09:49, 14.37s/it]
|
| 22 |
29%|βββ | 16/56 [03:59<09:23, 14.08s/it]
|
| 23 |
30%|βββ | 17/56 [04:11<08:50, 13.60s/it]
|
| 24 |
32%|ββββ | 18/56 [04:25<08:33, 13.52s/it]
|
| 25 |
34%|ββββ | 19/56 [04:38<08:21, 13.55s/it]
|
| 26 |
36%|ββββ | 20/56 [04:52<08:07, 13.55s/it]
|
| 27 |
|
|
|
|
| 28 |
36%|ββββ | 20/56 [04:52<08:07, 13.55s/it]
|
| 29 |
38%|ββββ | 21/56 [05:03<07:28, 12.80s/it]
|
| 30 |
39%|ββββ | 22/56 [05:14<06:51, 12.11s/it]
|
| 31 |
41%|ββββ | 23/56 [05:27<06:52, 12.51s/it]
|
| 32 |
43%|βββββ | 24/56 [05:41<06:55, 12.99s/it]
|
| 33 |
45%|βββββ | 25/56 [05:54<06:39, 12.90s/it]
|
| 34 |
|
|
|
|
| 35 |
45%|βββββ | 25/56 [05:54<06:39, 12.90s/it]
|
| 36 |
46%|βββββ | 26/56 [06:06<06:18, 12.61s/it]
|
| 37 |
48%|βββββ | 27/56 [06:18<05:59, 12.41s/it]
|
| 38 |
50%|βββββ | 28/56 [06:31<05:55, 12.71s/it]
|
| 39 |
52%|ββββββ | 29/56 [06:44<05:41, 12.65s/it]
|
| 40 |
54%|ββββββ | 30/56 [06:58<05:40, 13.11s/it]
|
| 41 |
|
|
|
|
| 42 |
54%|ββββββ | 30/56 [06:58<05:40, 13.11s/it]
|
| 43 |
55%|ββββββ | 31/56 [07:11<05:31, 13.24s/it]
|
| 44 |
57%|ββββββ | 32/56 [07:24<05:17, 13.22s/it]
|
| 45 |
59%|ββββββ | 33/56 [07:36<04:53, 12.76s/it]
|
| 46 |
61%|ββββββ | 34/56 [07:49<04:38, 12.67s/it]
|
| 47 |
62%|βββββββ | 35/56 [08:01<04:24, 12.61s/it]
|
| 48 |
|
|
|
|
| 49 |
62%|βββββββ | 35/56 [08:01<04:24, 12.61s/it]
|
| 50 |
64%|βββββββ | 36/56 [08:14<04:12, 12.63s/it]
|
| 51 |
66%|βββββββ | 37/56 [08:25<03:49, 12.10s/it]
|
| 52 |
68%|βββββββ | 38/56 [08:37<03:40, 12.27s/it]
|
| 53 |
70%|βββββββ | 39/56 [08:47<03:17, 11.61s/it]
|
| 54 |
71%|ββββββββ | 40/56 [09:03<03:24, 12.76s/it]
|
| 55 |
|
|
|
|
| 56 |
71%|ββββββββ | 40/56 [09:03<03:24, 12.76s/it]
|
| 57 |
73%|ββββββββ | 41/56 [09:18<03:22, 13.50s/it]
|
| 58 |
75%|ββββββββ | 42/56 [09:31<03:06, 13.32s/it]
|
| 59 |
77%|ββββββββ | 43/56 [09:45<02:55, 13.51s/it]
|
| 60 |
79%|ββββββββ | 44/56 [09:58<02:40, 13.40s/it]
|
| 61 |
80%|ββββββββ | 45/56 [10:12<02:29, 13.62s/it]
|
| 62 |
|
|
|
|
| 63 |
80%|ββββββββ | 45/56 [10:12<02:29, 13.62s/it]
|
| 64 |
82%|βββββββββ | 46/56 [10:23<02:07, 12.77s/it]
|
| 65 |
84%|βββββββββ | 47/56 [10:35<01:52, 12.49s/it]
|
| 66 |
86%|βββββββββ | 48/56 [10:47<01:38, 12.31s/it]
|
| 67 |
88%|βββββββββ | 49/56 [10:58<01:24, 12.04s/it]
|
| 68 |
89%|βββββββββ | 50/56 [11:12<01:16, 12.73s/it]
|
| 69 |
|
|
|
|
| 70 |
89%|βββββββββ | 50/56 [11:12<01:16, 12.73s/it]Unsloth: Restored added_tokens_decoder metadata in outputs/sft_adapter/checkpoint-50/tokenizer_config.json.
|
|
|
|
|
|
|
| 71 |
91%|βββββββββ | 51/56 [11:24<01:02, 12.53s/it]
|
| 72 |
93%|ββββββββββ| 52/56 [11:39<00:52, 13.04s/it]
|
| 73 |
95%|ββββββββββ| 53/56 [11:51<00:38, 12.90s/it]
|
| 74 |
96%|ββββββββββ| 54/56 [12:06<00:26, 13.38s/it]
|
| 75 |
98%|ββββββββββ| 55/56 [12:16<00:12, 12.42s/it]
|
| 76 |
|
|
|
|
| 77 |
98%|ββββββββββ| 55/56 [12:16<00:12, 12.42s/it]
|
|
|
|
| 78 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 79 |
...adapter/training_args.bin: 100%|ββββββββββ| 5.78kB / 5.78kB [A[A
|
|
|
|
|
|
|
|
|
|
| 80 |
...ft_adapter/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB [A[A[A
|
|
|
|
|
|
|
|
|
|
|
|
|
| 81 |
...adapter_model.safetensors: 84%|βββββββοΏ½οΏ½οΏ½β | 272MB / 323MB [A[A[A[A
|
|
|
|
|
|
|
| 82 |
...adapter/training_args.bin: 100%|ββββββββββ| 5.78kB / 5.78kB [A[A
|
|
|
|
|
|
|
|
|
|
| 83 |
...ft_adapter/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB [A[A[A
|
|
|
|
|
|
|
|
|
|
|
|
|
| 84 |
...adapter_model.safetensors: 84%|βββββββββ | 272MB / 323MB [A[A[A[A
|
|
|
|
|
|
|
| 85 |
...adapter/training_args.bin: 100%|ββββββββββ| 5.78kB / 5.78kB [A[A
|
|
|
|
|
|
|
|
|
|
| 86 |
...ft_adapter/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB [A[A[A
|
|
|
|
|
|
|
|
|
|
|
|
|
| 87 |
...adapter_model.safetensors: 100%|ββββββββββ| 323MB / 323MB [A[A[A[A
|
|
|
|
|
|
|
| 88 |
...adapter/training_args.bin: 100%|ββββββββββ| 5.78kB / 5.78kB [A[A
|
|
|
|
|
|
|
|
|
|
| 89 |
...ft_adapter/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB [A[A[A
|
|
|
|
|
|
|
|
|
|
|
|
|
| 90 |
...adapter_model.safetensors: 100%|ββββββββββ| 323MB / 323MB [A[A[A[A
|
|
|
|
|
|
|
| 91 |
...adapter/training_args.bin: 100%|ββββββββββ| 5.78kB / 5.78kB [A[A
|
|
|
|
|
|
|
|
|
|
| 92 |
...ft_adapter/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB [A[A[A
|
|
|
|
|
|
|
|
|
|
|
|
|
| 93 |
...adapter_model.safetensors: 100%|ββββββββββ| 323MB / 323MB [A[A[A[A
|
|
|
|
|
|
|
| 94 |
...adapter/training_args.bin: 100%|ββββββββββ| 5.78kB / 5.78kB
|
|
|
|
| 95 |
...ft_adapter/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB
|
|
|
|
| 96 |
...adapter_model.safetensors: 100%|ββββββββββ| 323MB / 323MB
|
|
|
|
|
|
|
|
|
|
|
|
|
| 97 |
...adapter/training_args.bin: 100%|ββββββββββ| 5.78kB / 5.78kB [A[A
|
|
|
|
|
|
|
|
|
|
| 98 |
...ft_adapter/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB [A[A[A
|
|
|
|
|
|
|
|
|
|
|
|
|
| 99 |
...adapter_model.safetensors: 82%|βββββββββ | 264MB / 323MB [A[A[A[A
|
|
|
|
|
|
|
| 100 |
...adapter/training_args.bin: 100%|ββββββββββ| 5.78kB / 5.78kB [A[A
|
|
|
|
|
|
|
|
|
|
| 101 |
...ft_adapter/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB [A[A[A
|
|
|
|
|
|
|
|
|
|
|
|
|
| 102 |
...adapter_model.safetensors: 82%|βββββββββ | 264MB / 323MB [A[A[A[A
|
|
|
|
|
|
|
| 103 |
...adapter/training_args.bin: 100%|ββββββββββ| 5.78kB / 5.78kB [A[A
|
|
|
|
|
|
|
|
|
|
| 104 |
...ft_adapter/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB [A[A[A
|
|
|
|
|
|
|
|
|
|
|
|
|
| 105 |
...adapter_model.safetensors: 100%|ββββββββββ| 323MB / 323MB [A[A[A[A
|
|
|
|
|
|
|
| 106 |
...adapter/training_args.bin: 100%|ββββββββββ| 5.78kB / 5.78kB [A[A
|
|
|
|
|
|
|
|
|
|
| 107 |
...ft_adapter/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB [A[A[A
|
|
|
|
|
|
|
|
|
|
|
|
|
| 108 |
...adapter_model.safetensors: 100%|ββββββββββ| 323MB / 323MB [A[A[A[A
|
|
|
|
|
|
|
| 109 |
...adapter/training_args.bin: 100%|ββββββββββ| 5.78kB / 5.78kB [A[A
|
|
|
|
|
|
|
|
|
|
| 110 |
...ft_adapter/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB [A[A[A
|
|
|
|
|
|
|
|
|
|
|
|
|
| 111 |
...adapter_model.safetensors: 100%|ββββββββββ| 323MB / 323MB [A[A[A[A
|
|
|
|
|
|
|
| 112 |
...adapter/training_args.bin: 100%|ββββββββββ| 5.78kB / 5.78kB
|
|
|
|
| 113 |
...ft_adapter/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB
|
|
|
|
| 114 |
...adapter_model.safetensors: 100%|ββββββββββ| 323MB / 323MB
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
pid: 108652
|
| 3 |
+
cwd: "/teamspace/studios/this_studio/corp_gym"
|
| 4 |
+
command: "stdbuf -oL -eL python -u training/train_sft.py \\\n --model Qwen/Qwen2.5-7B-Instruct \\\n --data data/sft/e1_m1_h1_examples.jsonl \\\n --output outputs/sft_adapter \\\n --epochs 2 \\\n --batch-size 1 --grad-accum 8 \\\n --max-seq-length 8192 \\\n --dataloader-num-workers 2 \\\n --dataset-num-proc 4 \\\n --legacy-text \\\n --push-to-hub Navigam/corp-env-sft-qwen2.5-7b 2>&1 | tee results/runs/sft_train.log"
|
| 5 |
+
started_at: 2026-04-25T16:22:09.902Z
|
| 6 |
+
running_for_ms: 795016
|
| 7 |
+
---
|
| 8 |
+
π¦₯ Unsloth: Will patch your computer to enable 2x faster free finetuning.
|
| 9 |
+
π¦₯ Unsloth Zoo will now patch everything to make training faster!
|
| 10 |
+
==((====))== Unsloth 2026.4.8: Fast Qwen2 patching. Transformers: 5.5.0.
|
| 11 |
+
\\ /| NVIDIA H100 80GB HBM3. Num GPUs = 1. Max memory: 79.189 GB. Platform: Linux.
|
| 12 |
+
O^O/ \_/ \ Torch: 2.7.1+cu128. CUDA: 9.0. CUDA Toolkit: 12.8. Triton: 3.3.1
|
| 13 |
+
\ / Bfloat16 = TRUE. FA [Xformers = 0.0.31.post1. FA2 = True]
|
| 14 |
+
"-____-" Free license: http://github.com/unslothai/unsloth
|
| 15 |
+
Unsloth: Fast downloading is enabled - ignore downloading bars which are red colored!
|
| 16 |
+
|
| 17 |
+
unsloth/qwen2.5-7b-instruct-unsloth-bnb-4bit does not have a padding token! Will use pad_token = <|PAD_TOKEN|>.
|
| 18 |
+
Unsloth 2026.4.8 patched 28 layers with 28 QKV layers, 28 O layers and 28 MLP layers.
|
| 19 |
+
warmup_ratio is deprecated and will be removed in v5.2. Use `warmup_steps` instead.
|
| 20 |
+
|
| 21 |
+
π¦₯ Unsloth: Padding-free auto-enabled, enabling faster training.
|
| 22 |
+
The tokenizer has new PAD/BOS/EOS tokens that differ from the model config and generation config. The model config and generation config were aligned accordingly, being updated with the tokenizer's values. Updated tokens: {'bos_token_id': None}.
|
| 23 |
+
==((====))== Unsloth - 2x faster free finetuning | Num GPUs used = 1
|
| 24 |
+
\\ /| Num examples = 224 | Num Epochs = 2 | Total steps = 56
|
| 25 |
+
O^O/ \_/ \ Batch size per device = 1 | Gradient accumulation steps = 8
|
| 26 |
+
\ / Data Parallel GPUs = 1 | Total batch size (1 x 8 x 1) = 8
|
| 27 |
+
"-____-" Trainable parameters = 80,740,352 of 7,696,356,864 (1.05% trained)
|
| 28 |
+
|
| 29 |
0%| | 0/56 [00:00<?, ?it/s]`use_return_dict` is deprecated! Use `return_dict` instead!
|
| 30 |
+
Unsloth: Will smartly offload gradients to save VRAM!
|
| 31 |
+
|
| 32 |
2%|β | 1/56 [00:11<10:36, 11.57s/it]
|
| 33 |
4%|β | 2/56 [00:23<10:39, 11.85s/it]
|
| 34 |
5%|β | 3/56 [00:38<11:40, 13.21s/it]
|
| 35 |
7%|β | 4/56 [00:51<11:27, 13.21s/it]
|
| 36 |
9%|β | 5/56 [01:07<11:53, 13.99s/it]
|
| 37 |
|
| 38 |
+
|
| 39 |
9%|β | 5/56 [01:07<11:53, 13.99s/it]
|
| 40 |
11%|β | 6/56 [01:25<12:48, 15.37s/it]
|
| 41 |
12%|ββ | 7/56 [01:40<12:31, 15.34s/it]
|
| 42 |
14%|ββ | 8/56 [01:59<13:06, 16.39s/it]
|
| 43 |
16%|ββ | 9/56 [02:18<13:40, 17.47s/it]
|
| 44 |
18%|ββ | 10/56 [02:33<12:45, 16.64s/it]
|
| 45 |
|
| 46 |
+
|
| 47 |
18%|ββ | 10/56 [02:33<12:45, 16.64s/it]
|
| 48 |
20%|ββ | 11/56 [02:52<12:56, 17.25s/it]
|
| 49 |
21%|βββ | 12/56 [03:07<12:18, 16.78s/it]
|
| 50 |
23%|βββ | 13/56 [03:18<10:43, 14.96s/it]
|
| 51 |
25%|βββ | 14/56 [03:31<10:02, 14.34s/it]
|
| 52 |
27%|βββ | 15/56 [03:46<09:49, 14.37s/it]
|
| 53 |
|
| 54 |
+
|
| 55 |
27%|βββ | 15/56 [03:46<09:49, 14.37s/it]
|
| 56 |
29%|βββ | 16/56 [03:59<09:23, 14.08s/it]
|
| 57 |
30%|βββ | 17/56 [04:11<08:50, 13.60s/it]
|
| 58 |
32%|ββββ | 18/56 [04:25<08:33, 13.52s/it]
|
| 59 |
34%|ββββ | 19/56 [04:38<08:21, 13.55s/it]
|
| 60 |
36%|ββββ | 20/56 [04:52<08:07, 13.55s/it]
|
| 61 |
|
| 62 |
+
|
| 63 |
36%|ββββ | 20/56 [04:52<08:07, 13.55s/it]
|
| 64 |
38%|ββββ | 21/56 [05:03<07:28, 12.80s/it]
|
| 65 |
39%|ββββ | 22/56 [05:14<06:51, 12.11s/it]
|
| 66 |
41%|ββββ | 23/56 [05:27<06:52, 12.51s/it]
|
| 67 |
43%|βββββ | 24/56 [05:41<06:55, 12.99s/it]
|
| 68 |
45%|βββββ | 25/56 [05:54<06:39, 12.90s/it]
|
| 69 |
|
| 70 |
+
|
| 71 |
45%|βββββ | 25/56 [05:54<06:39, 12.90s/it]
|
| 72 |
46%|βββββ | 26/56 [06:06<06:18, 12.61s/it]
|
| 73 |
48%|βββββ | 27/56 [06:18<05:59, 12.41s/it]
|
| 74 |
50%|βββββ | 28/56 [06:31<05:55, 12.71s/it]
|
| 75 |
52%|ββββββ | 29/56 [06:44<05:41, 12.65s/it]
|
| 76 |
54%|ββββββ | 30/56 [06:58<05:40, 13.11s/it]
|
| 77 |
|
| 78 |
+
|
| 79 |
54%|ββββββ | 30/56 [06:58<05:40, 13.11s/it]
|
| 80 |
55%|ββββββ | 31/56 [07:11<05:31, 13.24s/it]
|
| 81 |
57%|ββββββ | 32/56 [07:24<05:17, 13.22s/it]
|
| 82 |
59%|ββββββ | 33/56 [07:36<04:53, 12.76s/it]
|
| 83 |
61%|ββββββ | 34/56 [07:49<04:38, 12.67s/it]
|
| 84 |
62%|βββββββ | 35/56 [08:01<04:24, 12.61s/it]
|
| 85 |
|
| 86 |
+
|
| 87 |
62%|βββββββ | 35/56 [08:01<04:24, 12.61s/it]
|
| 88 |
64%|βββββββ | 36/56 [08:14<04:12, 12.63s/it]
|
| 89 |
66%|βββββββ | 37/56 [08:25<03:49, 12.10s/it]
|
| 90 |
68%|βββββββ | 38/56 [08:37<03:40, 12.27s/it]
|
| 91 |
70%|βββββββ | 39/56 [08:47<03:17, 11.61s/it]
|
| 92 |
71%|ββββββββ | 40/56 [09:03<03:24, 12.76s/it]
|
| 93 |
|
| 94 |
+
|
| 95 |
71%|ββββββββ | 40/56 [09:03<03:24, 12.76s/it]
|
| 96 |
73%|ββββββββ | 41/56 [09:18<03:22, 13.50s/it]
|
| 97 |
75%|ββββββββ | 42/56 [09:31<03:06, 13.32s/it]
|
| 98 |
77%|ββββββββ | 43/56 [09:45<02:55, 13.51s/it]
|
| 99 |
79%|ββββββββ | 44/56 [09:58<02:40, 13.40s/it]
|
| 100 |
80%|ββββββββ | 45/56 [10:12<02:29, 13.62s/it]
|
| 101 |
|
| 102 |
+
|
| 103 |
80%|ββββββββ | 45/56 [10:12<02:29, 13.62s/it]
|
| 104 |
82%|βββββββββ | 46/56 [10:23<02:07, 12.77s/it]
|
| 105 |
84%|βββββββββ | 47/56 [10:35<01:52, 12.49s/it]
|
| 106 |
86%|βββββββββ | 48/56 [10:47<01:38, 12.31s/it]
|
| 107 |
88%|βββββββββ | 49/56 [10:58<01:24, 12.04s/it]
|
| 108 |
89%|βββββββββ | 50/56 [11:12<01:16, 12.73s/it]
|
| 109 |
|
| 110 |
+
|
| 111 |
89%|βββββββββ | 50/56 [11:12<01:16, 12.73s/it]Unsloth: Restored added_tokens_decoder metadata in outputs/sft_adapter/checkpoint-50/tokenizer_config.json.
|
| 112 |
+
Unsloth: Restored added_tokens_decoder metadata in outputs/sft_adapter/tokenizer_config.json.
|
| 113 |
+
|
| 114 |
91%|βββββββββ | 51/56 [11:24<01:02, 12.53s/it]
|
| 115 |
93%|ββββββββββ| 52/56 [11:39<00:52, 13.04s/it]
|
| 116 |
95%|ββββββββββ| 53/56 [11:51<00:38, 12.90s/it]
|
| 117 |
96%|ββββββββββ| 54/56 [12:06<00:26, 13.38s/it]
|
| 118 |
98%|ββββββββββ| 55/56 [12:16<00:12, 12.42s/it]
|
| 119 |
|
| 120 |
+
|
| 121 |
98%|ββββββββββ| 55/56 [12:16<00:12, 12.42s/it]
|
| 122 |
+
|
| 123 |
|
| 124 |
+
|
| 125 |
+
|
| 126 |
+
|
| 127 |
+
|
| 128 |
+
|
| 129 |
...adapter/training_args.bin: 100%|ββββββββββ| 5.78kB / 5.78kB [A[A
|
| 130 |
+
|
| 131 |
+
|
| 132 |
+
|
| 133 |
...ft_adapter/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB [A[A[A
|
| 134 |
+
|
| 135 |
+
|
| 136 |
+
|
| 137 |
+
|
| 138 |
...adapter_model.safetensors: 84%|βββββββοΏ½οΏ½οΏ½β | 272MB / 323MB [A[A[A[A
|
| 139 |
+
|
| 140 |
+
|
| 141 |
...adapter/training_args.bin: 100%|ββββββββββ| 5.78kB / 5.78kB [A[A
|
| 142 |
+
|
| 143 |
+
|
| 144 |
+
|
| 145 |
...ft_adapter/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB [A[A[A
|
| 146 |
+
|
| 147 |
+
|
| 148 |
+
|
| 149 |
+
|
| 150 |
...adapter_model.safetensors: 84%|βββββββββ | 272MB / 323MB [A[A[A[A
|
| 151 |
+
|
| 152 |
+
|
| 153 |
...adapter/training_args.bin: 100%|ββββββββββ| 5.78kB / 5.78kB [A[A
|
| 154 |
+
|
| 155 |
+
|
| 156 |
+
|
| 157 |
...ft_adapter/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB [A[A[A
|
| 158 |
+
|
| 159 |
+
|
| 160 |
+
|
| 161 |
+
|
| 162 |
...adapter_model.safetensors: 100%|ββββββββββ| 323MB / 323MB [A[A[A[A
|
| 163 |
+
|
| 164 |
+
|
| 165 |
...adapter/training_args.bin: 100%|ββββββββββ| 5.78kB / 5.78kB [A[A
|
| 166 |
+
|
| 167 |
+
|
| 168 |
+
|
| 169 |
...ft_adapter/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB [A[A[A
|
| 170 |
+
|
| 171 |
+
|
| 172 |
+
|
| 173 |
+
|
| 174 |
...adapter_model.safetensors: 100%|ββββββββββ| 323MB / 323MB [A[A[A[A
|
| 175 |
+
|
| 176 |
+
|
| 177 |
...adapter/training_args.bin: 100%|ββββββββββ| 5.78kB / 5.78kB [A[A
|
| 178 |
+
|
| 179 |
+
|
| 180 |
+
|
| 181 |
...ft_adapter/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB [A[A[A
|
| 182 |
+
|
| 183 |
+
|
| 184 |
+
|
| 185 |
+
|
| 186 |
...adapter_model.safetensors: 100%|ββββββββββ| 323MB / 323MB [A[A[A[A
|
| 187 |
+
|
| 188 |
+
|
| 189 |
...adapter/training_args.bin: 100%|ββββββββββ| 5.78kB / 5.78kB
|
| 190 |
+
|
| 191 |
...ft_adapter/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB
|
| 192 |
+
|
| 193 |
...adapter_model.safetensors: 100%|ββββββββββ| 323MB / 323MB
|
| 194 |
+
|
| 195 |
+
|
| 196 |
+
|
| 197 |
+
|
| 198 |
...adapter/training_args.bin: 100%|ββββββββββ| 5.78kB / 5.78kB [A[A
|
| 199 |
+
|
| 200 |
+
|
| 201 |
+
|
| 202 |
...ft_adapter/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB [A[A[A
|
| 203 |
+
|
| 204 |
+
|
| 205 |
+
|
| 206 |
+
|
| 207 |
...adapter_model.safetensors: 82%|βββββββββ | 264MB / 323MB [A[A[A[A
|
| 208 |
+
|
| 209 |
+
|
| 210 |
...adapter/training_args.bin: 100%|ββββββββββ| 5.78kB / 5.78kB [A[A
|
| 211 |
+
|
| 212 |
+
|
| 213 |
+
|
| 214 |
...ft_adapter/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB [A[A[A
|
| 215 |
+
|
| 216 |
+
|
| 217 |
+
|
| 218 |
+
|
| 219 |
...adapter_model.safetensors: 82%|βββββββββ | 264MB / 323MB [A[A[A[A
|
| 220 |
+
|
| 221 |
+
|
| 222 |
...adapter/training_args.bin: 100%|ββββββββββ| 5.78kB / 5.78kB [A[A
|
| 223 |
+
|
| 224 |
+
|
| 225 |
+
|
| 226 |
...ft_adapter/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB [A[A[A
|
| 227 |
+
|
| 228 |
+
|
| 229 |
+
|
| 230 |
+
|
| 231 |
...adapter_model.safetensors: 100%|ββββββββββ| 323MB / 323MB [A[A[A[A
|
| 232 |
+
|
| 233 |
+
|
| 234 |
...adapter/training_args.bin: 100%|ββββββββββ| 5.78kB / 5.78kB [A[A
|
| 235 |
+
|
| 236 |
+
|
| 237 |
+
|
| 238 |
...ft_adapter/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB [A[A[A
|
| 239 |
+
|
| 240 |
+
|
| 241 |
+
|
| 242 |
+
|
| 243 |
...adapter_model.safetensors: 100%|ββββββββββ| 323MB / 323MB [A[A[A[A
|
| 244 |
+
|
| 245 |
+
|
| 246 |
...adapter/training_args.bin: 100%|ββββββββββ| 5.78kB / 5.78kB [A[A
|
| 247 |
+
|
| 248 |
+
|
| 249 |
+
|
| 250 |
...ft_adapter/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB [A[A[A
|
| 251 |
+
|
| 252 |
+
|
| 253 |
+
|
| 254 |
+
|
| 255 |
...adapter_model.safetensors: 100%|ββββββββββ| 323MB / 323MB [A[A[A[A
|
| 256 |
+
|
| 257 |
+
|
| 258 |
...adapter/training_args.bin: 100%|ββββββββββ| 5.78kB / 5.78kB
|
| 259 |
+
|
| 260 |
...ft_adapter/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB
|
| 261 |
+
|
| 262 |
...adapter_model.safetensors: 100%|ββββββββββ| 323MB / 323MB
|
| 263 |
+
No files have been modified since last commit. Skipping to prevent empty commit.
|
| 264 |
+
[huggingface_hub.hf_api|WARNING]No files have been modified since last commit. Skipping to prevent empty commit.
|
| 265 |
+
|
| 266 |
+
---
|
| 267 |
+
exit_code: 0
|
| 268 |
+
elapsed_ms: 796210
|
| 269 |
+
ended_at: 2026-04-25T16:35:26.112Z
|
| 270 |
+
---
|
training_logs/386081.txt
ADDED
|
@@ -0,0 +1,47 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 0 |
0%| | 0/50 [00:00<?, ?it/s]Unsloth: Will smartly offload gradients to save VRAM!
|
|
|
|
| 1 |
2%|β | 1/50 [00:40<33:22, 40.86s/it]
|
| 2 |
4%|β | 2/50 [01:15<29:44, 37.18s/it]
|
| 3 |
6%|β | 3/50 [01:52<29:05, 37.13s/it]
|
| 4 |
8%|β | 4/50 [02:32<29:17, 38.21s/it]
|
| 5 |
10%|β | 5/50 [03:06<27:38, 36.86s/it]
|
| 6 |
|
|
|
|
| 7 |
10%|β | 5/50 [03:06<27:38, 36.86s/it]
|
| 8 |
12%|ββ | 6/50 [03:38<25:46, 35.15s/it]
|
| 9 |
14%|ββ | 7/50 [04:21<26:58, 37.65s/it]
|
| 10 |
16%|ββ | 8/50 [04:53<25:01, 35.74s/it]
|
| 11 |
18%|ββ | 9/50 [05:30<24:44, 36.22s/it]
|
| 12 |
20%|ββ | 10/50 [06:10<24:57, 37.43s/it]
|
| 13 |
|
|
|
|
| 14 |
20%|ββ | 10/50 [06:10<24:57, 37.43s/it]
|
| 15 |
22%|βββ | 11/50 [06:50<24:48, 38.17s/it]
|
| 16 |
24%|βββ | 12/50 [07:22<22:58, 36.27s/it]
|
| 17 |
26%|βββ | 13/50 [07:56<22:03, 35.76s/it]
|
| 18 |
28%|βββ | 14/50 [08:31<21:12, 35.33s/it]
|
| 19 |
30%|βββ | 15/50 [09:16<22:21, 38.33s/it]
|
| 20 |
|
|
|
|
| 21 |
30%|βββ | 15/50 [09:16<22:21, 38.33s/it]
|
| 22 |
32%|ββββ | 16/50 [09:56<21:59, 38.80s/it]
|
| 23 |
34%|ββββ | 17/50 [10:36<21:33, 39.21s/it]
|
| 24 |
36%|ββββ | 18/50 [11:13<20:35, 38.60s/it]
|
| 25 |
38%|ββββ | 19/50 [11:48<19:18, 37.36s/it]
|
| 26 |
40%|ββββ | 20/50 [12:36<20:15, 40.52s/it]
|
| 27 |
|
|
|
|
| 28 |
40%|ββββ | 20/50 [12:36<20:15, 40.52s/it]
|
| 29 |
42%|βββββ | 21/50 [13:10<18:44, 38.77s/it]
|
| 30 |
44%|βββββ | 22/50 [13:45<17:27, 37.43s/it]
|
| 31 |
46%|βββββ | 23/50 [14:16<16:04, 35.73s/it]
|
| 32 |
48%|βββββ | 24/50 [14:56<16:01, 37.00s/it]
|
| 33 |
50%|βββββ | 25/50 [15:39<16:06, 38.67s/it]
|
| 34 |
|
|
|
|
| 35 |
50%|βββββ | 25/50 [15:39<16:06, 38.67s/it]
|
| 36 |
52%|ββββββ | 26/50 [16:13<14:57, 37.39s/it]
|
| 37 |
54%|ββββββ | 27/50 [16:56<14:57, 39.01s/it]
|
| 38 |
56%|ββββββ | 28/50 [17:36<14:24, 39.27s/it]
|
| 39 |
58%|ββββββ | 29/50 [18:13<13:31, 38.64s/it]
|
| 40 |
60%|ββββββ | 30/50 [18:48<12:28, 37.40s/it]
|
| 41 |
|
|
|
|
| 42 |
60%|ββββββ | 30/50 [18:48<12:28, 37.40s/it]
|
| 43 |
62%|βββββββ | 31/50 [19:28<12:04, 38.15s/it]
|
| 44 |
64%|βββββββ | 32/50 [19:57<10:38, 35.49s/it]
|
| 45 |
66%|βββββββ | 33/50 [20:34<10:12, 36.02s/it]
|
| 46 |
68%|βββββββ | 34/50 [21:08<09:28, 35.53s/it]
|
| 47 |
70%|βββββββ | 35/50 [21:48<09:12, 36.83s/it]
|
| 48 |
|
|
|
|
| 49 |
70%|βββββββ | 35/50 [21:48<09:12, 36.83s/it]
|
| 50 |
72%|ββββββββ | 36/50 [22:26<08:37, 36.98s/it]
|
| 51 |
74%|ββββββββ | 37/50 [23:03<08:01, 37.02s/it]
|
| 52 |
76%|ββββββββ | 38/50 [23:40<07:26, 37.20s/it]
|
| 53 |
78%|ββββββββ | 39/50 [24:20<06:57, 37.99s/it]
|
| 54 |
80%|ββββββββ | 40/50 [24:52<06:01, 36.12s/it]
|
| 55 |
|
|
|
|
| 56 |
80%|ββββββββ | 40/50 [24:52<06:01, 36.12s/it]
|
| 57 |
82%|βββββββββ | 41/50 [25:32<05:34, 37.22s/it]
|
| 58 |
84%|βββββββββ | 42/50 [26:09<04:57, 37.19s/it]
|
| 59 |
86%|βββββββββ | 43/50 [26:54<04:37, 39.61s/it]
|
| 60 |
88%|βββββββββ | 44/50 [27:31<03:53, 38.90s/it]
|
| 61 |
90%|βββββββββ | 45/50 [28:11<03:15, 39.19s/it]
|
| 62 |
|
|
|
|
| 63 |
90%|βββββββββ | 45/50 [28:11<03:15, 39.19s/it]
|
| 64 |
92%|ββββββββββ| 46/50 [28:46<02:31, 37.84s/it]
|
| 65 |
94%|ββββββββββ| 47/50 [29:21<01:50, 36.96s/it]
|
| 66 |
96%|ββββββββββ| 48/50 [29:58<01:14, 37.03s/it]
|
| 67 |
98%|ββββββββββ| 49/50 [30:35<00:37, 37.09s/it]
|
| 68 |
|
|
|
|
|
|
|
| 69 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
pid: 123381
|
| 3 |
+
cwd: "/teamspace/studios/this_studio/corp_gym"
|
| 4 |
+
command: "cd /teamspace/studios/this_studio/corp_gym && python3 training/train_sft.py \\\n --model Qwen/Qwen3.5-9B \\\n --data data/sft/e1_m1_examples.jsonl \\\n --output outputs/sft_Qwen3.5-9B \\\n --max-steps -1 \\\n 2>&1 | tee /tmp/train_qwen35.log | tail -35"
|
| 5 |
+
started_at: 2026-04-25T12:36:16.644Z
|
| 6 |
+
running_for_ms: 1936277
|
| 7 |
+
---
|
| 8 |
+
π¦₯ Unsloth: Will patch your computer to enable 2x faster free finetuning.
|
| 9 |
+
Unsloth: Your Flash Attention 2 installation seems to be broken. Using Xformers instead. No performance changes will be seen.
|
| 10 |
+
π¦₯ Unsloth Zoo will now patch everything to make training faster!
|
| 11 |
+
==((====))== Unsloth 2026.4.8: Fast Qwen3_5 patching. Transformers: 5.5.0.
|
| 12 |
+
\\ /| NVIDIA H100 80GB HBM3. Num GPUs = 1. Max memory: 79.096 GB. Platform: Linux.
|
| 13 |
+
O^O/ \_/ \ Torch: 2.10.0+cu128. CUDA: 9.0. CUDA Toolkit: 12.8. Triton: 3.6.0
|
| 14 |
+
\ / Bfloat16 = TRUE. FA [Xformers = 0.0.35. FA2 = False]
|
| 15 |
+
"-____-" Free license: http://github.com/unslothai/unsloth
|
| 16 |
+
Unsloth: Fast downloading is enabled - ignore downloading bars which are red colored!
|
| 17 |
+
The fast path is not available because one of the required library is not installed. Falling back to torch implementation. To install follow https://github.com/fla-org/flash-linear-attention#installation and https://github.com/Dao-AILab/causal-conv1d
|
| 18 |
+
|
| 19 |
+
warmup_ratio is deprecated and will be removed in v5.2. Use `warmup_steps` instead.
|
| 20 |
+
|
| 21 |
+
The tokenizer has new PAD/BOS/EOS tokens that differ from the model config and generation config. The model config and generation config were aligned accordingly, being updated with the tokenizer's values. Updated tokens: {'eos_token_id': 248046}.
|
| 22 |
+
==((====))== Unsloth - 2x faster free finetuning | Num GPUs used = 1
|
| 23 |
+
\\ /| Num examples = 200 | Num Epochs = 2 | Total steps = 50
|
| 24 |
+
O^O/ \_/ \ Batch size per device = 1 | Gradient accumulation steps = 8
|
| 25 |
+
\ / Data Parallel GPUs = 1 | Total batch size (1 x 8 x 1) = 8
|
| 26 |
+
"-____-" Trainable parameters = 58,195,968 of 9,468,009,712 (0.61% trained)
|
| 27 |
+
|
| 28 |
0%| | 0/50 [00:00<?, ?it/s]Unsloth: Will smartly offload gradients to save VRAM!
|
| 29 |
+
|
| 30 |
2%|β | 1/50 [00:40<33:22, 40.86s/it]
|
| 31 |
4%|β | 2/50 [01:15<29:44, 37.18s/it]
|
| 32 |
6%|β | 3/50 [01:52<29:05, 37.13s/it]
|
| 33 |
8%|β | 4/50 [02:32<29:17, 38.21s/it]
|
| 34 |
10%|β | 5/50 [03:06<27:38, 36.86s/it]
|
| 35 |
|
| 36 |
+
|
| 37 |
10%|β | 5/50 [03:06<27:38, 36.86s/it]
|
| 38 |
12%|ββ | 6/50 [03:38<25:46, 35.15s/it]
|
| 39 |
14%|ββ | 7/50 [04:21<26:58, 37.65s/it]
|
| 40 |
16%|ββ | 8/50 [04:53<25:01, 35.74s/it]
|
| 41 |
18%|ββ | 9/50 [05:30<24:44, 36.22s/it]
|
| 42 |
20%|ββ | 10/50 [06:10<24:57, 37.43s/it]
|
| 43 |
|
| 44 |
+
|
| 45 |
20%|ββ | 10/50 [06:10<24:57, 37.43s/it]
|
| 46 |
22%|βββ | 11/50 [06:50<24:48, 38.17s/it]
|
| 47 |
24%|βββ | 12/50 [07:22<22:58, 36.27s/it]
|
| 48 |
26%|βββ | 13/50 [07:56<22:03, 35.76s/it]
|
| 49 |
28%|βββ | 14/50 [08:31<21:12, 35.33s/it]
|
| 50 |
30%|βββ | 15/50 [09:16<22:21, 38.33s/it]
|
| 51 |
|
| 52 |
+
|
| 53 |
30%|βββ | 15/50 [09:16<22:21, 38.33s/it]
|
| 54 |
32%|ββββ | 16/50 [09:56<21:59, 38.80s/it]
|
| 55 |
34%|ββββ | 17/50 [10:36<21:33, 39.21s/it]
|
| 56 |
36%|ββββ | 18/50 [11:13<20:35, 38.60s/it]
|
| 57 |
38%|ββββ | 19/50 [11:48<19:18, 37.36s/it]
|
| 58 |
40%|ββββ | 20/50 [12:36<20:15, 40.52s/it]
|
| 59 |
|
| 60 |
+
|
| 61 |
40%|ββββ | 20/50 [12:36<20:15, 40.52s/it]
|
| 62 |
42%|βββββ | 21/50 [13:10<18:44, 38.77s/it]
|
| 63 |
44%|βββββ | 22/50 [13:45<17:27, 37.43s/it]
|
| 64 |
46%|βββββ | 23/50 [14:16<16:04, 35.73s/it]
|
| 65 |
48%|βββββ | 24/50 [14:56<16:01, 37.00s/it]
|
| 66 |
50%|βββββ | 25/50 [15:39<16:06, 38.67s/it]
|
| 67 |
|
| 68 |
+
|
| 69 |
50%|βββββ | 25/50 [15:39<16:06, 38.67s/it]
|
| 70 |
52%|ββββββ | 26/50 [16:13<14:57, 37.39s/it]
|
| 71 |
54%|ββββββ | 27/50 [16:56<14:57, 39.01s/it]
|
| 72 |
56%|ββββββ | 28/50 [17:36<14:24, 39.27s/it]
|
| 73 |
58%|ββββββ | 29/50 [18:13<13:31, 38.64s/it]
|
| 74 |
60%|ββββββ | 30/50 [18:48<12:28, 37.40s/it]
|
| 75 |
|
| 76 |
+
|
| 77 |
60%|ββββββ | 30/50 [18:48<12:28, 37.40s/it]
|
| 78 |
62%|βββββββ | 31/50 [19:28<12:04, 38.15s/it]
|
| 79 |
64%|βββββββ | 32/50 [19:57<10:38, 35.49s/it]
|
| 80 |
66%|βββββββ | 33/50 [20:34<10:12, 36.02s/it]
|
| 81 |
68%|βββββββ | 34/50 [21:08<09:28, 35.53s/it]
|
| 82 |
70%|βββββββ | 35/50 [21:48<09:12, 36.83s/it]
|
| 83 |
|
| 84 |
+
|
| 85 |
70%|βββββββ | 35/50 [21:48<09:12, 36.83s/it]
|
| 86 |
72%|ββββββββ | 36/50 [22:26<08:37, 36.98s/it]
|
| 87 |
74%|ββββββββ | 37/50 [23:03<08:01, 37.02s/it]
|
| 88 |
76%|ββββββββ | 38/50 [23:40<07:26, 37.20s/it]
|
| 89 |
78%|ββββββββ | 39/50 [24:20<06:57, 37.99s/it]
|
| 90 |
80%|ββββββββ | 40/50 [24:52<06:01, 36.12s/it]
|
| 91 |
|
| 92 |
+
|
| 93 |
80%|ββββββββ | 40/50 [24:52<06:01, 36.12s/it]
|
| 94 |
82%|βββββββββ | 41/50 [25:32<05:34, 37.22s/it]
|
| 95 |
84%|βββββββββ | 42/50 [26:09<04:57, 37.19s/it]
|
| 96 |
86%|βββββββββ | 43/50 [26:54<04:37, 39.61s/it]
|
| 97 |
88%|βββββββββ | 44/50 [27:31<03:53, 38.90s/it]
|
| 98 |
90%|βββββββββ | 45/50 [28:11<03:15, 39.19s/it]
|
| 99 |
|
| 100 |
+
|
| 101 |
90%|βββββββββ | 45/50 [28:11<03:15, 39.19s/it]
|
| 102 |
92%|ββββββββββ| 46/50 [28:46<02:31, 37.84s/it]
|
| 103 |
94%|ββββββββββ| 47/50 [29:21<01:50, 36.96s/it]
|
| 104 |
96%|ββββββββββ| 48/50 [29:58<01:14, 37.03s/it]
|
| 105 |
98%|ββββββββββ| 49/50 [30:35<00:37, 37.09s/it]
|
| 106 |
|
| 107 |
+
|
| 108 |
+
|
| 109 |
|
| 110 |
+
|
| 111 |
+
Unsloth: Restored added_tokens_decoder metadata in outputs/sft_Qwen3.5-9B/tokenizer_config.json.
|
| 112 |
+
|
| 113 |
+
---
|
| 114 |
+
exit_code: 0
|
| 115 |
+
elapsed_ms: 1936836
|
| 116 |
+
ended_at: 2026-04-25T13:08:33.480Z
|
| 117 |
+
---
|
training_logs/432673.txt
ADDED
|
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
pid: 277866
|
| 3 |
+
cwd: "/teamspace/studios/this_studio/corp_gym"
|
| 4 |
+
command: "cd /teamspace/studios/this_studio/corp_gym && python3 eval.py --policy hf \\\n --model Qwen/Qwen2.5-7B-Instruct \\\n --adapter outputs/grpo_Qwen2.5-7B-Instruct \\\n --label grpo-qwen2.5-7b 2>&1"
|
| 5 |
+
started_at: 2026-04-25T14:24:41.899Z
|
| 6 |
+
running_for_ms: 345001
|
| 7 |
+
---
|
| 8 |
+
Skipping import of cpp extensions due to incompatible torch version. Please upgrade to torch >= 2.11.0 (found 2.10.0+cu128).
|
| 9 |
+
`torch_dtype` is deprecated! Use `dtype` instead!
|
| 10 |
+
|
| 11 |
+
The following generation flags are not valid and may be ignored: ['temperature', 'top_p', 'top_k']. Set `TRANSFORMERS_VERBOSITY=info` for more details.
|
| 12 |
+
This is a friendly reminder - the current text generation call has exceeded the model's predefined maximum length (32768). Depending on the model, you may observe exceptions, performance degradation, or nothing at all.
|
| 13 |
+
|
| 14 |
+
[grpo-qwen2.5-7b] 3 episodes
|
| 15 |
+
- e1_launch_readiness: terminal_reward=0.910 pass_rate=1.000 success=1.000
|
| 16 |
+
- m1_budget_reallocation: terminal_reward=0.548 pass_rate=0.500 success=0.000
|
| 17 |
+
- h1_acquisition_defence: terminal_reward=-0.010 pass_rate=0.417 success=0.000
|
| 18 |
+
|
| 19 |
+
Wrote results/runs/Qwen2.5-7B-Instruct__grpo_Qwen2.5-7B-Instruct__grpo-qwen2.5-7b/grpo-qwen2.5-7b_eval.jsonl
|
| 20 |
+
Wrote results/runs/Qwen2.5-7B-Instruct__grpo_Qwen2.5-7B-Instruct__grpo-qwen2.5-7b/metadata.json
|
| 21 |
+
|
| 22 |
+
---
|
| 23 |
+
exit_code: 0
|
| 24 |
+
elapsed_ms: 346053
|
| 25 |
+
ended_at: 2026-04-25T14:30:27.952Z
|
| 26 |
+
---
|
training_logs/524754.txt
ADDED
|
@@ -0,0 +1,207 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 0 |
0%| | 0/200 [00:00<?, ?it/s]`use_return_dict` is deprecated! Use `return_dict` instead!
|
|
|
|
|
|
|
| 1 |
0%| | 1/200 [00:12<40:59, 12.36s/it]
|
| 2 |
1%| | 2/200 [00:19<30:03, 9.11s/it]
|
| 3 |
2%|β | 3/200 [00:25<26:26, 8.05s/it]
|
| 4 |
2%|β | 4/200 [00:32<24:42, 7.56s/it]
|
| 5 |
2%|β | 5/200 [00:39<23:43, 7.30s/it]
|
| 6 |
|
| 7 |
2%|β | 5/200 [00:39<23:43, 7.30s/it]{'loss': '0.9696', 'grad_norm': '0.2933', 'learning_rate': '8e-05', 'epoch': '0.1786'}
|
|
|
|
| 8 |
3%|β | 6/200 [00:46<23:11, 7.17s/it]
|
| 9 |
4%|β | 7/200 [00:53<22:41, 7.06s/it]
|
| 10 |
4%|β | 8/200 [01:00<22:24, 7.00s/it]
|
| 11 |
4%|β | 9/200 [01:07<22:16, 7.00s/it]
|
| 12 |
5%|β | 10/200 [01:14<21:59, 6.94s/it]
|
| 13 |
|
|
|
|
| 14 |
5%|β | 10/200 [01:14<21:59, 6.94s/it]
|
| 15 |
6%|β | 11/200 [01:21<21:57, 6.97s/it]
|
| 16 |
6%|β | 12/200 [01:27<21:45, 6.94s/it]
|
| 17 |
6%|β | 13/200 [01:34<21:25, 6.87s/it]
|
| 18 |
7%|β | 14/200 [01:41<21:14, 6.85s/it]
|
| 19 |
8%|β | 15/200 [01:48<21:12, 6.88s/it]
|
| 20 |
|
|
|
|
| 21 |
8%|β | 15/200 [01:48<21:12, 6.88s/it]
|
| 22 |
8%|β | 16/200 [01:55<21:05, 6.88s/it]
|
| 23 |
8%|β | 17/200 [02:02<20:56, 6.86s/it]
|
| 24 |
9%|β | 18/200 [02:08<20:46, 6.85s/it]
|
| 25 |
10%|β | 19/200 [02:15<20:41, 6.86s/it]
|
| 26 |
10%|β | 20/200 [02:22<20:31, 6.84s/it]
|
| 27 |
|
|
|
|
| 28 |
10%|β | 20/200 [02:22<20:31, 6.84s/it]
|
| 29 |
10%|β | 21/200 [02:29<20:21, 6.83s/it]
|
| 30 |
11%|β | 22/200 [02:36<20:07, 6.78s/it]
|
| 31 |
12%|ββ | 23/200 [02:42<20:04, 6.81s/it]
|
| 32 |
12%|ββ | 24/200 [02:49<20:04, 6.84s/it]
|
| 33 |
12%|ββ | 25/200 [02:56<19:56, 6.84s/it]
|
| 34 |
|
| 35 |
12%|ββ | 25/200 [02:56<19:56, 6.84s/it]{'loss': '0.079', 'grad_norm': '0.08526', 'learning_rate': '0.0001973', 'epoch': '0.8929'}
|
|
|
|
| 36 |
13%|ββ | 26/200 [03:03<19:49, 6.84s/it]
|
| 37 |
14%|ββ | 27/200 [03:10<19:42, 6.84s/it]
|
| 38 |
14%|ββ | 28/200 [03:17<19:39, 6.86s/it]
|
| 39 |
14%|ββ | 29/200 [03:24<19:35, 6.87s/it]
|
| 40 |
15%|ββ | 30/200 [03:31<19:30, 6.89s/it]
|
| 41 |
|
|
|
|
| 42 |
15%|ββ | 30/200 [03:31<19:30, 6.89s/it]
|
| 43 |
16%|ββ | 31/200 [03:38<19:26, 6.90s/it]
|
| 44 |
16%|ββ | 32/200 [03:44<19:19, 6.90s/it]
|
| 45 |
16%|ββ | 33/200 [03:51<19:05, 6.86s/it]
|
| 46 |
17%|ββ | 34/200 [03:58<18:57, 6.85s/it]
|
| 47 |
18%|ββ | 35/200 [04:05<18:49, 6.85s/it]
|
| 48 |
|
|
|
|
| 49 |
18%|ββ | 35/200 [04:05<18:49, 6.85s/it]
|
| 50 |
18%|ββ | 36/200 [04:12<18:46, 6.87s/it]
|
| 51 |
18%|ββ | 37/200 [04:19<18:32, 6.83s/it]
|
| 52 |
19%|ββ | 38/200 [04:25<18:27, 6.83s/it]
|
| 53 |
20%|ββ | 39/200 [04:32<18:14, 6.80s/it]
|
| 54 |
20%|ββ | 40/200 [04:39<18:16, 6.86s/it]
|
| 55 |
|
| 56 |
20%|ββ | 40/200 [04:39<18:16, 6.86s/it]{'loss': '0.04835', 'grad_norm': '0.03554', 'learning_rate': '0.0001887', 'epoch': '1.429'}
|
|
|
|
| 57 |
20%|ββ | 41/200 [04:46<18:16, 6.89s/it]
|
| 58 |
21%|ββ | 42/200 [04:53<18:04, 6.86s/it]
|
| 59 |
22%|βββ | 43/200 [05:00<18:01, 6.89s/it]
|
| 60 |
22%|βββ | 44/200 [05:07<17:54, 6.89s/it]
|
| 61 |
22%|βββ | 45/200 [05:14<17:50, 6.90s/it]
|
| 62 |
|
| 63 |
22%|βββ | 45/200 [05:14<17:50, 6.90s/it]{'loss': '0.04462', 'grad_norm': '0.03563', 'learning_rate': '0.0001846', 'epoch': '1.607'}
|
|
|
|
| 64 |
23%|βββ | 46/200 [05:20<17:37, 6.87s/it]
|
| 65 |
24%|βββ | 47/200 [05:27<17:29, 6.86s/it]
|
| 66 |
24%|βββ | 48/200 [05:34<17:21, 6.85s/it]
|
| 67 |
24%|βββ | 49/200 [05:41<17:10, 6.82s/it]
|
| 68 |
25%|βββ | 50/200 [05:48<17:05, 6.84s/it]
|
| 69 |
|
|
|
|
| 70 |
25%|βββ | 50/200 [05:48<17:05, 6.84s/it]Unsloth: Restored added_tokens_decoder metadata in outputs/sft_deepseek_14b/checkpoint-50/tokenizer_config.json.
|
|
|
|
|
|
|
| 71 |
26%|βββ | 51/200 [05:56<18:22, 7.40s/it]
|
| 72 |
26%|βββ | 52/200 [06:03<17:54, 7.26s/it]
|
| 73 |
26%|βββ | 53/200 [06:10<17:30, 7.15s/it]
|
| 74 |
27%|βββ | 54/200 [06:17<17:14, 7.08s/it]
|
| 75 |
28%|βββ | 55/200 [06:24<16:51, 6.98s/it]
|
| 76 |
|
|
|
|
| 77 |
28%|βββ | 55/200 [06:24<16:51, 6.98s/it]
|
| 78 |
28%|βββ | 56/200 [06:31<16:38, 6.93s/it]
|
| 79 |
28%|βββ | 57/200 [06:38<16:33, 6.95s/it]
|
| 80 |
29%|βββ | 58/200 [06:45<16:22, 6.92s/it]
|
| 81 |
30%|βββ | 59/200 [06:51<16:11, 6.89s/it]
|
| 82 |
30%|βββ | 60/200 [06:58<16:07, 6.91s/it]
|
| 83 |
|
|
|
|
| 84 |
30%|βββ | 60/200 [06:58<16:07, 6.91s/it]
|
| 85 |
30%|βββ | 61/200 [07:05<15:58, 6.90s/it]
|
| 86 |
31%|βββ | 62/200 [07:12<15:51, 6.90s/it]
|
| 87 |
32%|ββββ | 63/200 [07:19<15:39, 6.86s/it]
|
| 88 |
32%|ββββ | 64/200 [07:26<15:29, 6.84s/it]
|
| 89 |
32%|ββββ | 65/200 [07:32<15:17, 6.80s/it]
|
| 90 |
|
|
|
|
| 91 |
32%|ββββ | 65/200 [07:32<15:17, 6.80s/it]
|
| 92 |
33%|ββββ | 66/200 [07:39<15:11, 6.80s/it]
|
| 93 |
34%|ββββ | 67/200 [07:46<15:04, 6.80s/it]
|
| 94 |
34%|ββββ | 68/200 [07:53<14:56, 6.79s/it]
|
| 95 |
34%|ββββ | 69/200 [08:00<14:53, 6.82s/it]
|
| 96 |
35%|ββββ | 70/200 [08:07<14:48, 6.84s/it]
|
| 97 |
|
| 98 |
35%|ββββ | 70/200 [08:07<14:48, 6.84s/it]{'loss': '0.04121', 'grad_norm': '0.04701', 'learning_rate': '0.0001561', 'epoch': '2.5'}
|
|
|
|
| 99 |
36%|ββββ | 71/200 [08:13<14:42, 6.84s/it]
|
| 100 |
36%|ββββ | 72/200 [08:20<14:36, 6.85s/it]
|
| 101 |
36%|ββββ | 73/200 [08:27<14:33, 6.88s/it]
|
| 102 |
37%|ββββ | 74/200 [08:34<14:31, 6.91s/it]
|
| 103 |
38%|ββββ | 75/200 [08:41<14:24, 6.92s/it]
|
| 104 |
|
|
|
|
| 105 |
38%|ββββ | 75/200 [08:41<14:24, 6.92s/it]
|
| 106 |
38%|ββββ | 76/200 [08:48<14:16, 6.91s/it]
|
| 107 |
38%|ββββ | 77/200 [08:55<14:10, 6.92s/it]
|
| 108 |
39%|ββββ | 78/200 [09:02<13:58, 6.87s/it]
|
| 109 |
40%|ββββ | 79/200 [09:09<13:51, 6.88s/it]
|
| 110 |
40%|ββββ | 80/200 [09:15<13:43, 6.86s/it]
|
| 111 |
|
|
|
|
| 112 |
40%|ββββ | 80/200 [09:15<13:43, 6.86s/it]
|
| 113 |
40%|ββββ | 81/200 [09:22<13:36, 6.86s/it]
|
| 114 |
41%|ββββ | 82/200 [09:29<13:32, 6.89s/it]
|
| 115 |
42%|βββββ | 83/200 [09:36<13:22, 6.86s/it]
|
| 116 |
42%|βββββ | 84/200 [09:43<13:11, 6.82s/it]
|
| 117 |
42%|βββββ | 85/200 [09:50<13:09, 6.87s/it]
|
| 118 |
|
|
|
|
| 119 |
42%|βββββ | 85/200 [09:50<13:09, 6.87s/it]
|
| 120 |
43%|βββββ | 86/200 [09:57<13:03, 6.87s/it]
|
| 121 |
44%|βββββ | 87/200 [10:03<12:56, 6.87s/it]
|
| 122 |
44%|βββββ | 88/200 [10:10<12:49, 6.87s/it]
|
| 123 |
44%|βββββ | 89/200 [10:17<12:37, 6.83s/it]
|
| 124 |
45%|βββββ | 90/200 [10:24<12:31, 6.83s/it]
|
| 125 |
|
|
|
|
| 126 |
45%|βββββ | 90/200 [10:24<12:31, 6.83s/it]
|
| 127 |
46%|βββββ | 91/200 [10:31<12:27, 6.86s/it]
|
| 128 |
46%|βββββ | 92/200 [10:38<12:22, 6.87s/it]
|
| 129 |
46%|βββββ | 93/200 [10:45<12:12, 6.85s/it]
|
| 130 |
47%|βββββ | 94/200 [10:51<12:03, 6.82s/it]
|
| 131 |
48%|βββββ | 95/200 [10:58<11:56, 6.82s/it]
|
| 132 |
|
| 133 |
48%|βββββ | 95/200 [10:58<11:56, 6.82s/it]{'loss': '0.03869', 'grad_norm': '0.01334', 'learning_rate': '0.0001181', 'epoch': '3.393'}
|
|
|
|
| 134 |
48%|βββββ | 96/200 [11:05<11:52, 6.85s/it]
|
| 135 |
48%|βββββ | 97/200 [11:12<11:50, 6.90s/it]
|
| 136 |
49%|βββββ | 98/200 [11:19<11:45, 6.91s/it]
|
| 137 |
50%|βββββ | 99/200 [11:26<11:36, 6.89s/it]
|
| 138 |
50%|βββββ | 100/200 [11:33<11:27, 6.87s/it]
|
| 139 |
|
|
|
|
| 140 |
50%|βββββ | 100/200 [11:33<11:27, 6.87s/it]Unsloth: Restored added_tokens_decoder metadata in outputs/sft_deepseek_14b/checkpoint-100/tokenizer_config.json.
|
|
|
|
| 141 |
50%|βββββ | 101/200 [11:41<12:11, 7.39s/it]
|
| 142 |
51%|βββββ | 102/200 [11:48<11:49, 7.24s/it]
|
| 143 |
52%|ββββββ | 103/200 [11:55<11:30, 7.12s/it]
|
| 144 |
52%|ββββββ | 104/200 [12:02<11:15, 7.03s/it]
|
| 145 |
52%|ββββββ | 105/200 [12:09<11:03, 6.99s/it]
|
| 146 |
|
|
|
|
| 147 |
52%|ββββββ | 105/200 [12:09<11:03, 6.99s/it]
|
| 148 |
53%|ββββββ | 106/200 [12:16<10:53, 6.95s/it]
|
| 149 |
54%|ββββββ | 107/200 [12:22<10:42, 6.91s/it]
|
| 150 |
54%|ββββββ | 108/200 [12:29<10:31, 6.86s/it]
|
| 151 |
55%|ββββββ | 109/200 [12:36<10:20, 6.82s/it]
|
| 152 |
55%|ββββββ | 110/200 [12:43<10:12, 6.80s/it]
|
| 153 |
|
|
|
|
| 154 |
55%|ββββββ | 110/200 [12:43<10:12, 6.80s/it]
|
| 155 |
56%|ββββββ | 111/200 [12:49<10:07, 6.82s/it]
|
| 156 |
56%|ββββββ | 112/200 [12:56<10:04, 6.86s/it]
|
| 157 |
56%|ββββββ | 113/200 [13:04<10:09, 7.00s/it]
|
| 158 |
57%|ββββββ | 114/200 [13:11<09:59, 6.97s/it]
|
| 159 |
57%|ββββββ | 115/200 [13:18<09:51, 6.96s/it]
|
| 160 |
|
|
|
|
| 161 |
57%|ββββββ | 115/200 [13:18<09:51, 6.96s/it]
|
| 162 |
58%|ββββββ | 116/200 [13:25<09:43, 6.95s/it]
|
| 163 |
58%|ββββββ | 117/200 [13:31<09:31, 6.88s/it]
|
| 164 |
59%|ββββββ | 118/200 [13:38<09:24, 6.88s/it]
|
| 165 |
60%|ββββββ | 119/200 [13:45<09:18, 6.90s/it]
|
| 166 |
60%|ββββββ | 120/200 [13:52<09:12, 6.91s/it]
|
| 167 |
|
| 168 |
60%|ββββββ | 120/200 [13:52<09:12, 6.91s/it]{'loss': '0.03772', 'grad_norm': '0.01068', 'learning_rate': '7.706e-05', 'epoch': '4.286'}
|
|
|
|
| 169 |
60%|ββββββ | 121/200 [13:59<09:00, 6.84s/it]
|
| 170 |
61%|ββββββ | 122/200 [14:05<08:52, 6.83s/it]
|
| 171 |
62%|βββββββ | 123/200 [14:12<08:45, 6.82s/it]
|
| 172 |
62%|βββββββ | 124/200 [14:19<08:36, 6.80s/it]
|
| 173 |
62%|βββββββ | 125/200 [14:26<08:31, 6.83s/it]
|
| 174 |
|
| 175 |
62%|βββββββ | 125/200 [14:26<08:31, 6.83s/it]{'loss': '0.039', 'grad_norm': '0.01277', 'learning_rate': '6.91e-05', 'epoch': '4.464'}
|
|
|
|
| 176 |
63%|βββββββ | 126/200 [14:33<08:25, 6.83s/it]
|
| 177 |
64%|βββββββ | 127/200 [14:40<08:18, 6.83s/it]
|
| 178 |
64%|βββββββ | 128/200 [14:46<08:09, 6.79s/it]
|
| 179 |
64%|βββββββ | 129/200 [14:53<08:04, 6.82s/it]
|
| 180 |
65%|βββββββ | 130/200 [15:00<08:00, 6.86s/it]
|
| 181 |
|
|
|
|
| 182 |
65%|βββββββ | 130/200 [15:00<08:00, 6.86s/it]
|
| 183 |
66%|βββββββ | 131/200 [15:07<07:53, 6.87s/it]
|
| 184 |
66%|βββββββ | 132/200 [15:14<07:48, 6.89s/it]
|
| 185 |
66%|βββββββ | 133/200 [15:21<07:38, 6.84s/it]
|
| 186 |
67%|βββββββ | 134/200 [15:27<07:28, 6.80s/it]
|
| 187 |
68%|βββββββ | 135/200 [15:34<07:23, 6.82s/it]
|
| 188 |
|
| 189 |
68%|βββββββ | 135/200 [15:34<07:23, 6.82s/it]{'loss': '0.03844', 'grad_norm': '0.01827', 'learning_rate': '5.387e-05', 'epoch': '4.821'}
|
|
|
|
| 190 |
68%|βββββββ | 136/200 [15:41<07:19, 6.87s/it]
|
| 191 |
68%|βββββββ | 137/200 [15:48<07:14, 6.90s/it]
|
| 192 |
69%|βββββββ | 138/200 [15:55<07:07, 6.89s/it]
|
| 193 |
70%|βββββββ | 139/200 [16:02<06:58, 6.86s/it]
|
| 194 |
70%|βββββββ | 140/200 [16:09<06:52, 6.88s/it]
|
| 195 |
|
| 196 |
70%|βββββββ | 140/200 [16:09<06:52, 6.88s/it]{'loss': '0.03757', 'grad_norm': '0.01235', 'learning_rate': '4.67e-05', 'epoch': '5'}
|
|
|
|
| 197 |
70%|βββββββ | 141/200 [16:16<06:44, 6.85s/it]
|
| 198 |
71%|βββββββ | 142/200 [16:23<06:39, 6.90s/it]
|
| 199 |
72%|ββββββββ | 143/200 [16:29<06:31, 6.86s/it]
|
| 200 |
72%|ββββββββ | 144/200 [16:36<06:23, 6.85s/it]
|
| 201 |
72%|ββββββββ | 145/200 [16:43<06:17, 6.87s/it]
|
| 202 |
|
|
|
|
| 203 |
72%|ββββββββ | 145/200 [16:43<06:17, 6.87s/it]
|
| 204 |
73%|ββββββββ | 146/200 [16:50<06:11, 6.88s/it]
|
| 205 |
74%|ββββββββ | 147/200 [16:57<06:06, 6.91s/it]
|
| 206 |
74%|ββββββββ | 148/200 [17:04<06:00, 6.93s/it]
|
| 207 |
74%|ββββββββ | 149/200 [17:11<05:52, 6.91s/it]
|
| 208 |
75%|ββββββββ | 150/200 [17:18<05:44, 6.88s/it]
|
| 209 |
|
|
|
|
| 210 |
75%|ββββββββ | 150/200 [17:18<05:44, 6.88s/it]Unsloth: Restored added_tokens_decoder metadata in outputs/sft_deepseek_14b/checkpoint-150/tokenizer_config.json.
|
|
|
|
| 211 |
76%|ββββββββ | 151/200 [17:26<06:02, 7.41s/it]
|
| 212 |
76%|ββββββββ | 152/200 [17:33<05:47, 7.23s/it]
|
| 213 |
76%|ββββββββ | 153/200 [17:40<05:34, 7.11s/it]
|
| 214 |
77%|ββββββββ | 154/200 [17:47<05:26, 7.09s/it]
|
| 215 |
78%|ββββββββ | 155/200 [17:54<05:16, 7.03s/it]
|
| 216 |
|
|
|
|
| 217 |
78%|ββββββββ | 155/200 [17:54<05:16, 7.03s/it]
|
| 218 |
78%|ββββββββ | 156/200 [18:01<05:06, 6.97s/it]
|
| 219 |
78%|ββββββββ | 157/200 [18:07<04:57, 6.91s/it]
|
| 220 |
79%|ββββββββ | 158/200 [18:14<04:48, 6.87s/it]
|
| 221 |
80%|ββββββββ | 159/200 [18:21<04:41, 6.88s/it]
|
| 222 |
80%|ββββββββ | 160/200 [18:28<04:34, 6.87s/it]
|
| 223 |
|
|
|
|
| 224 |
80%|ββββββββ | 160/200 [18:28<04:34, 6.87s/it]
|
| 225 |
80%|ββββββοΏ½οΏ½β | 161/200 [18:35<04:26, 6.83s/it]
|
| 226 |
81%|ββββββββ | 162/200 [18:41<04:18, 6.81s/it]
|
| 227 |
82%|βββββββββ | 163/200 [18:48<04:12, 6.81s/it]
|
| 228 |
82%|βββββββββ | 164/200 [18:55<04:05, 6.83s/it]
|
| 229 |
82%|βββββββββ | 165/200 [19:02<03:59, 6.84s/it]
|
| 230 |
|
|
|
|
| 231 |
82%|βββββββββ | 165/200 [19:02<03:59, 6.84s/it]
|
| 232 |
83%|βββββββββ | 166/200 [19:09<03:53, 6.86s/it]
|
| 233 |
84%|βββββββββ | 167/200 [19:16<03:46, 6.87s/it]
|
| 234 |
84%|βββββββββ | 168/200 [19:23<03:38, 6.82s/it]
|
| 235 |
84%|βββββββββ | 169/200 [19:29<03:31, 6.83s/it]
|
| 236 |
85%|βββββββββ | 170/200 [19:36<03:24, 6.82s/it]
|
| 237 |
|
| 238 |
85%|βββββββββ | 170/200 [19:36<03:24, 6.82s/it]{'loss': '0.03837', 'grad_norm': '0.01728', 'learning_rate': '1.285e-05', 'epoch': '6.071'}
|
|
|
|
| 239 |
86%|βββββββββ | 171/200 [19:43<03:18, 6.86s/it]
|
| 240 |
86%|βββββββββ | 172/200 [19:50<03:11, 6.86s/it]
|
| 241 |
86%|βββββββββ | 173/200 [19:57<03:04, 6.84s/it]
|
| 242 |
87%|βββββββββ | 174/200 [20:04<02:58, 6.87s/it]
|
| 243 |
88%|βββββββββ | 175/200 [20:11<02:51, 6.86s/it]
|
| 244 |
|
|
|
|
| 245 |
88%|βββββββββ | 175/200 [20:11<02:51, 6.86s/it]
|
| 246 |
88%|βββββββββ | 176/200 [20:17<02:44, 6.86s/it]
|
| 247 |
88%|βββββββββ | 177/200 [20:24<02:38, 6.91s/it]
|
| 248 |
89%|βββββββββ | 178/200 [20:31<02:32, 6.91s/it]
|
| 249 |
90%|βββββββββ | 179/200 [20:38<02:25, 6.92s/it]
|
| 250 |
90%|βββββββββ | 180/200 [20:45<02:17, 6.87s/it]
|
| 251 |
|
| 252 |
90%|βββββββββ | 180/200 [20:45<02:17, 6.87s/it]{'loss': '0.03702', 'grad_norm': '0.01512', 'learning_rate': '5.968e-06', 'epoch': '6.429'}
|
|
|
|
| 253 |
90%|βββββββββ | 181/200 [20:52<02:11, 6.90s/it]
|
| 254 |
91%|βββββββββ | 182/200 [20:59<02:03, 6.85s/it]
|
| 255 |
92%|ββββββββββ| 183/200 [21:06<01:56, 6.84s/it]
|
| 256 |
92%|ββββββββββ| 184/200 [21:13<01:49, 6.86s/it]
|
| 257 |
92%|ββββββββββ| 185/200 [21:19<01:43, 6.88s/it]
|
| 258 |
|
|
|
|
| 259 |
92%|ββββββββββ| 185/200 [21:19<01:43, 6.88s/it]
|
| 260 |
93%|ββββββββββ| 186/200 [21:26<01:36, 6.88s/it]
|
| 261 |
94%|ββββββββββ| 187/200 [21:33<01:29, 6.91s/it]
|
| 262 |
94%|ββββββββββ| 188/200 [21:40<01:22, 6.83s/it]
|
| 263 |
94%|ββββββββββ| 189/200 [21:47<01:15, 6.85s/it]
|
| 264 |
95%|ββββββββββ| 190/200 [21:54<01:08, 6.85s/it]
|
| 265 |
|
| 266 |
95%|ββββββββββ| 190/200 [21:54<01:08, 6.85s/it]{'loss': '0.03777', 'grad_norm': '0.01418', 'learning_rate': '1.649e-06', 'epoch': '6.786'}
|
|
|
|
| 267 |
96%|ββββββββββ| 191/200 [22:01<01:01, 6.87s/it]
|
| 268 |
96%|ββββββββββ| 192/200 [22:07<00:54, 6.87s/it]
|
| 269 |
96%|ββββββββββ| 193/200 [22:14<00:47, 6.84s/it]
|
| 270 |
97%|ββββββββββ| 194/200 [22:21<00:40, 6.78s/it]
|
| 271 |
98%|ββββββββββ| 195/200 [22:28<00:33, 6.78s/it]
|
| 272 |
|
|
|
|
| 273 |
98%|ββββββββββ| 195/200 [22:28<00:33, 6.78s/it]
|
| 274 |
98%|ββββββββββ| 196/200 [22:35<00:27, 6.82s/it]
|
| 275 |
98%|ββββββββββ| 197/200 [22:42<00:20, 6.86s/it]
|
| 276 |
99%|ββββββββββ| 198/200 [22:48<00:13, 6.88s/it]
|
| 277 |
|
|
|
|
|
|
|
| 278 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 279 |
...eek_14b/training_args.bin: 100%|ββββββββββ| 5.78kB / 5.78kB [A[A
|
|
|
|
|
|
|
|
|
|
| 280 |
...epseek_14b/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB [A[A[A
|
|
|
|
|
|
|
|
|
|
|
|
|
| 281 |
...adapter_model.safetensors: 90%|βββββββββ | 248MB / 275MB [A[A[A[A
|
|
|
|
|
|
|
| 282 |
...eek_14b/training_args.bin: 100%|ββββββββββ| 5.78kB / 5.78kB [A[A
|
|
|
|
|
|
|
|
|
|
| 283 |
...epseek_14b/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB [A[A[A
|
|
|
|
|
|
|
|
|
|
|
|
|
| 284 |
...adapter_model.safetensors: 90%|βββββββββ | 248MB / 275MB [A[A[A[A
|
|
|
|
|
|
|
| 285 |
...eek_14b/training_args.bin: 100%|ββββββββββ| 5.78kB / 5.78kB [A[A
|
|
|
|
|
|
|
|
|
|
| 286 |
...epseek_14b/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB [A[A[A
|
|
|
|
|
|
|
|
|
|
|
|
|
| 287 |
...adapter_model.safetensors: 100%|ββββββββββ| 275MB / 275MB [A[A[A[A
|
|
|
|
|
|
|
| 288 |
...eek_14b/training_args.bin: 100%|ββββββββββ| 5.78kB / 5.78kB [A[A
|
|
|
|
|
|
|
|
|
|
| 289 |
...epseek_14b/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB [A[A[A
|
|
|
|
|
|
|
|
|
|
|
|
|
| 290 |
...adapter_model.safetensors: 100%|ββββββββββ| 275MB / 275MB [A[A[A[A
|
|
|
|
|
|
|
| 291 |
...eek_14b/training_args.bin: 100%|ββββββββββ| 5.78kB / 5.78kB [A[A
|
|
|
|
|
|
|
|
|
|
| 292 |
...epseek_14b/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB [A[A[A
|
|
|
|
|
|
|
|
|
|
|
|
|
| 293 |
...adapter_model.safetensors: 100%|ββββββββββ| 275MB / 275MB [A[A[A[A
|
|
|
|
|
|
|
| 294 |
...eek_14b/training_args.bin: 100%|ββββββββββ| 5.78kB / 5.78kB [A[A
|
|
|
|
|
|
|
|
|
|
| 295 |
...epseek_14b/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB [A[A[A
|
|
|
|
|
|
|
|
|
|
|
|
|
| 296 |
...adapter_model.safetensors: 100%|ββββββββββ| 275MB / 275MB [A[A[A[A
|
|
|
|
|
|
|
| 297 |
...eek_14b/training_args.bin: 100%|ββββββββββ| 5.78kB / 5.78kB
|
|
|
|
| 298 |
...epseek_14b/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB
|
|
|
|
| 299 |
...adapter_model.safetensors: 100%|ββββββββββ| 275MB / 275MB
|
|
|
|
|
|
|
|
|
|
|
|
|
| 300 |
...eek_14b/training_args.bin: 100%|ββββββββββ| 5.78kB / 5.78kB [A[A
|
|
|
|
|
|
|
|
|
|
| 301 |
...epseek_14b/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB [A[A[A
|
|
|
|
|
|
|
|
|
|
|
|
|
| 302 |
...adapter_model.safetensors: 96%|ββββββββββ| 264MB / 275MB [A[A[A[A
|
|
|
|
|
|
|
| 303 |
...eek_14b/training_args.bin: 100%|ββββββββββ| 5.78kB / 5.78kB [A[A
|
|
|
|
|
|
|
|
|
|
| 304 |
...epseek_14b/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB [A[A[A
|
|
|
|
|
|
|
|
|
|
|
|
|
| 305 |
...adapter_model.safetensors: 96%|ββββββββββ| 264MB / 275MB [A[A[A[A
|
|
|
|
|
|
|
| 306 |
...eek_14b/training_args.bin: 100%|ββββββββββ| 5.78kB / 5.78kB [A[A
|
|
|
|
|
|
|
|
|
|
| 307 |
...epseek_14b/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB [A[A[A
|
|
|
|
|
|
|
|
|
|
|
|
|
| 308 |
...adapter_model.safetensors: 100%|ββββββββββ| 275MB / 275MB [A[A[A[A
|
|
|
|
|
|
|
| 309 |
...eek_14b/training_args.bin: 100%|ββββββββββ| 5.78kB / 5.78kB [A[A
|
|
|
|
|
|
|
|
|
|
| 310 |
...epseek_14b/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB [A[A[A
|
|
|
|
|
|
|
|
|
|
|
|
|
| 311 |
...adapter_model.safetensors: 100%|ββββββββββ| 275MB / 275MB [A[A[A[A
|
|
|
|
|
|
|
| 312 |
...eek_14b/training_args.bin: 100%|ββββββββββ| 5.78kB / 5.78kB [A[A
|
|
|
|
|
|
|
|
|
|
| 313 |
...epseek_14b/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB [A[A[A
|
|
|
|
|
|
|
|
|
|
|
|
|
| 314 |
...adapter_model.safetensors: 100%|ββββββββββ| 275MB / 275MB [A[A[A[A
|
|
|
|
|
|
|
| 315 |
...eek_14b/training_args.bin: 100%|ββββββββββ| 5.78kB / 5.78kB [A[A
|
|
|
|
|
|
|
|
|
|
| 316 |
...epseek_14b/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB [A[A[A
|
|
|
|
|
|
|
|
|
|
|
|
|
| 317 |
...adapter_model.safetensors: 100%|ββββββββββ| 275MB / 275MB [A[A[A[A
|
|
|
|
|
|
|
| 318 |
...eek_14b/training_args.bin: 100%|ββββββββββ| 5.78kB / 5.78kB
|
|
|
|
| 319 |
...epseek_14b/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB
|
|
|
|
| 320 |
...adapter_model.safetensors: 100%|ββββββββββ| 275MB / 275MB
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
pid: 22326
|
| 3 |
+
cwd: "/teamspace/studios/this_studio/corp_gym"
|
| 4 |
+
command: "python -u training/train_sft_genral.py --model deepseek-ai/DeepSeek-R1-Distill-Qwen-14B --dataset-repo Navigam/corp-env-data --dataset-file data/sft/e1_m1_h1_examples.jsonl --output outputs/sft_deepseek_14b --hf-user Navigam --epochs 2 --max-steps 200 --batch-size 1 --grad-accum 8"
|
| 5 |
+
started_at: 2026-04-26T02:57:08.378Z
|
| 6 |
+
running_for_ms: 1450038
|
| 7 |
+
---
|
| 8 |
+
π¦₯ Unsloth: Will patch your computer to enable 2x faster free finetuning.
|
| 9 |
+
π¦₯ Unsloth Zoo will now patch everything to make training faster!
|
| 10 |
+
Loaded dataset file from HF: /teamspace/studios/this_studio/.cache/huggingface/hub/datasets--Navigam--corp-env-data/snapshots/8f1399529014c706451a157100d131e2b63fc619/data/sft/e1_m1_h1_examples.jsonl
|
| 11 |
+
==((====))== Unsloth 2026.4.8: Fast Qwen2 patching. Transformers: 5.5.0.
|
| 12 |
+
\\ /| NVIDIA H100 80GB HBM3. Num GPUs = 1. Max memory: 79.096 GB. Platform: Linux.
|
| 13 |
+
O^O/ \_/ \ Torch: 2.7.1+cu128. CUDA: 9.0. CUDA Toolkit: 12.8. Triton: 3.3.1
|
| 14 |
+
\ / Bfloat16 = TRUE. FA [Xformers = 0.0.31.post1. FA2 = True]
|
| 15 |
+
"-____-" Free license: http://github.com/unslothai/unsloth
|
| 16 |
+
Unsloth: Fast downloading is enabled - ignore downloading bars which are red colored!
|
| 17 |
+
|
| 18 |
+
Unsloth: Will load unsloth/deepseek-r1-distill-qwen-14b-unsloth-bnb-4bit as a legacy tokenizer.
|
| 19 |
+
unsloth/deepseek-r1-distill-qwen-14b-unsloth-bnb-4bit does not have a padding token! Will use pad_token = <|PAD_TOKEN|>.
|
| 20 |
+
Unsloth 2026.4.8 patched 48 layers with 48 QKV layers, 48 O layers and 48 MLP layers.
|
| 21 |
+
warmup_ratio is deprecated and will be removed in v5.2. Use `warmup_steps` instead.
|
| 22 |
+
Unsloth: We found double BOS tokens - we shall remove one automatically.
|
| 23 |
+
|
| 24 |
+
π¦₯ Unsloth: Padding-free auto-enabled, enabling faster training.
|
| 25 |
+
==((====))== Unsloth - 2x faster free finetuning | Num GPUs used = 1
|
| 26 |
+
\\ /| Num examples = 224 | Num Epochs = 8 | Total steps = 200
|
| 27 |
+
O^O/ \_/ \ Batch size per device = 1 | Gradient accumulation steps = 8
|
| 28 |
+
\ / Data Parallel GPUs = 1 | Total batch size (1 x 8 x 1) = 8
|
| 29 |
+
"-____-" Trainable parameters = 68,812,800 of 14,838,846,464 (0.46% trained)
|
| 30 |
+
|
| 31 |
0%| | 0/200 [00:00<?, ?it/s]`use_return_dict` is deprecated! Use `return_dict` instead!
|
| 32 |
+
Unsloth: Will smartly offload gradients to save VRAM!
|
| 33 |
+
|
| 34 |
0%| | 1/200 [00:12<40:59, 12.36s/it]
|
| 35 |
1%| | 2/200 [00:19<30:03, 9.11s/it]
|
| 36 |
2%|β | 3/200 [00:25<26:26, 8.05s/it]
|
| 37 |
2%|β | 4/200 [00:32<24:42, 7.56s/it]
|
| 38 |
2%|β | 5/200 [00:39<23:43, 7.30s/it]
|
| 39 |
|
| 40 |
2%|β | 5/200 [00:39<23:43, 7.30s/it]{'loss': '0.9696', 'grad_norm': '0.2933', 'learning_rate': '8e-05', 'epoch': '0.1786'}
|
| 41 |
+
|
| 42 |
3%|β | 6/200 [00:46<23:11, 7.17s/it]
|
| 43 |
4%|β | 7/200 [00:53<22:41, 7.06s/it]
|
| 44 |
4%|β | 8/200 [01:00<22:24, 7.00s/it]
|
| 45 |
4%|β | 9/200 [01:07<22:16, 7.00s/it]
|
| 46 |
5%|β | 10/200 [01:14<21:59, 6.94s/it]
|
| 47 |
|
| 48 |
+
|
| 49 |
5%|β | 10/200 [01:14<21:59, 6.94s/it]
|
| 50 |
6%|β | 11/200 [01:21<21:57, 6.97s/it]
|
| 51 |
6%|β | 12/200 [01:27<21:45, 6.94s/it]
|
| 52 |
6%|β | 13/200 [01:34<21:25, 6.87s/it]
|
| 53 |
7%|β | 14/200 [01:41<21:14, 6.85s/it]
|
| 54 |
8%|β | 15/200 [01:48<21:12, 6.88s/it]
|
| 55 |
|
| 56 |
+
|
| 57 |
8%|β | 15/200 [01:48<21:12, 6.88s/it]
|
| 58 |
8%|β | 16/200 [01:55<21:05, 6.88s/it]
|
| 59 |
8%|β | 17/200 [02:02<20:56, 6.86s/it]
|
| 60 |
9%|β | 18/200 [02:08<20:46, 6.85s/it]
|
| 61 |
10%|β | 19/200 [02:15<20:41, 6.86s/it]
|
| 62 |
10%|β | 20/200 [02:22<20:31, 6.84s/it]
|
| 63 |
|
| 64 |
+
|
| 65 |
10%|β | 20/200 [02:22<20:31, 6.84s/it]
|
| 66 |
10%|β | 21/200 [02:29<20:21, 6.83s/it]
|
| 67 |
11%|β | 22/200 [02:36<20:07, 6.78s/it]
|
| 68 |
12%|ββ | 23/200 [02:42<20:04, 6.81s/it]
|
| 69 |
12%|ββ | 24/200 [02:49<20:04, 6.84s/it]
|
| 70 |
12%|ββ | 25/200 [02:56<19:56, 6.84s/it]
|
| 71 |
|
| 72 |
12%|ββ | 25/200 [02:56<19:56, 6.84s/it]{'loss': '0.079', 'grad_norm': '0.08526', 'learning_rate': '0.0001973', 'epoch': '0.8929'}
|
| 73 |
+
|
| 74 |
13%|ββ | 26/200 [03:03<19:49, 6.84s/it]
|
| 75 |
14%|ββ | 27/200 [03:10<19:42, 6.84s/it]
|
| 76 |
14%|ββ | 28/200 [03:17<19:39, 6.86s/it]
|
| 77 |
14%|ββ | 29/200 [03:24<19:35, 6.87s/it]
|
| 78 |
15%|ββ | 30/200 [03:31<19:30, 6.89s/it]
|
| 79 |
|
| 80 |
+
|
| 81 |
15%|ββ | 30/200 [03:31<19:30, 6.89s/it]
|
| 82 |
16%|ββ | 31/200 [03:38<19:26, 6.90s/it]
|
| 83 |
16%|ββ | 32/200 [03:44<19:19, 6.90s/it]
|
| 84 |
16%|ββ | 33/200 [03:51<19:05, 6.86s/it]
|
| 85 |
17%|ββ | 34/200 [03:58<18:57, 6.85s/it]
|
| 86 |
18%|ββ | 35/200 [04:05<18:49, 6.85s/it]
|
| 87 |
|
| 88 |
+
|
| 89 |
18%|ββ | 35/200 [04:05<18:49, 6.85s/it]
|
| 90 |
18%|ββ | 36/200 [04:12<18:46, 6.87s/it]
|
| 91 |
18%|ββ | 37/200 [04:19<18:32, 6.83s/it]
|
| 92 |
19%|ββ | 38/200 [04:25<18:27, 6.83s/it]
|
| 93 |
20%|ββ | 39/200 [04:32<18:14, 6.80s/it]
|
| 94 |
20%|ββ | 40/200 [04:39<18:16, 6.86s/it]
|
| 95 |
|
| 96 |
20%|ββ | 40/200 [04:39<18:16, 6.86s/it]{'loss': '0.04835', 'grad_norm': '0.03554', 'learning_rate': '0.0001887', 'epoch': '1.429'}
|
| 97 |
+
|
| 98 |
20%|ββ | 41/200 [04:46<18:16, 6.89s/it]
|
| 99 |
21%|ββ | 42/200 [04:53<18:04, 6.86s/it]
|
| 100 |
22%|βββ | 43/200 [05:00<18:01, 6.89s/it]
|
| 101 |
22%|βββ | 44/200 [05:07<17:54, 6.89s/it]
|
| 102 |
22%|βββ | 45/200 [05:14<17:50, 6.90s/it]
|
| 103 |
|
| 104 |
22%|βββ | 45/200 [05:14<17:50, 6.90s/it]{'loss': '0.04462', 'grad_norm': '0.03563', 'learning_rate': '0.0001846', 'epoch': '1.607'}
|
| 105 |
+
|
| 106 |
23%|βββ | 46/200 [05:20<17:37, 6.87s/it]
|
| 107 |
24%|βββ | 47/200 [05:27<17:29, 6.86s/it]
|
| 108 |
24%|βββ | 48/200 [05:34<17:21, 6.85s/it]
|
| 109 |
24%|βββ | 49/200 [05:41<17:10, 6.82s/it]
|
| 110 |
25%|βββ | 50/200 [05:48<17:05, 6.84s/it]
|
| 111 |
|
| 112 |
+
|
| 113 |
25%|βββ | 50/200 [05:48<17:05, 6.84s/it]Unsloth: Restored added_tokens_decoder metadata in outputs/sft_deepseek_14b/checkpoint-50/tokenizer_config.json.
|
| 114 |
+
Unsloth: Restored added_tokens_decoder metadata in outputs/sft_deepseek_14b/tokenizer_config.json.
|
| 115 |
+
|
| 116 |
26%|βββ | 51/200 [05:56<18:22, 7.40s/it]
|
| 117 |
26%|βββ | 52/200 [06:03<17:54, 7.26s/it]
|
| 118 |
26%|βββ | 53/200 [06:10<17:30, 7.15s/it]
|
| 119 |
27%|βββ | 54/200 [06:17<17:14, 7.08s/it]
|
| 120 |
28%|βββ | 55/200 [06:24<16:51, 6.98s/it]
|
| 121 |
|
| 122 |
+
|
| 123 |
28%|βββ | 55/200 [06:24<16:51, 6.98s/it]
|
| 124 |
28%|βββ | 56/200 [06:31<16:38, 6.93s/it]
|
| 125 |
28%|βββ | 57/200 [06:38<16:33, 6.95s/it]
|
| 126 |
29%|βββ | 58/200 [06:45<16:22, 6.92s/it]
|
| 127 |
30%|βββ | 59/200 [06:51<16:11, 6.89s/it]
|
| 128 |
30%|βββ | 60/200 [06:58<16:07, 6.91s/it]
|
| 129 |
|
| 130 |
+
|
| 131 |
30%|βββ | 60/200 [06:58<16:07, 6.91s/it]
|
| 132 |
30%|βββ | 61/200 [07:05<15:58, 6.90s/it]
|
| 133 |
31%|βββ | 62/200 [07:12<15:51, 6.90s/it]
|
| 134 |
32%|ββββ | 63/200 [07:19<15:39, 6.86s/it]
|
| 135 |
32%|ββββ | 64/200 [07:26<15:29, 6.84s/it]
|
| 136 |
32%|ββββ | 65/200 [07:32<15:17, 6.80s/it]
|
| 137 |
|
| 138 |
+
|
| 139 |
32%|ββββ | 65/200 [07:32<15:17, 6.80s/it]
|
| 140 |
33%|ββββ | 66/200 [07:39<15:11, 6.80s/it]
|
| 141 |
34%|ββββ | 67/200 [07:46<15:04, 6.80s/it]
|
| 142 |
34%|ββββ | 68/200 [07:53<14:56, 6.79s/it]
|
| 143 |
34%|ββββ | 69/200 [08:00<14:53, 6.82s/it]
|
| 144 |
35%|ββββ | 70/200 [08:07<14:48, 6.84s/it]
|
| 145 |
|
| 146 |
35%|ββββ | 70/200 [08:07<14:48, 6.84s/it]{'loss': '0.04121', 'grad_norm': '0.04701', 'learning_rate': '0.0001561', 'epoch': '2.5'}
|
| 147 |
+
|
| 148 |
36%|ββββ | 71/200 [08:13<14:42, 6.84s/it]
|
| 149 |
36%|ββββ | 72/200 [08:20<14:36, 6.85s/it]
|
| 150 |
36%|ββββ | 73/200 [08:27<14:33, 6.88s/it]
|
| 151 |
37%|ββββ | 74/200 [08:34<14:31, 6.91s/it]
|
| 152 |
38%|ββββ | 75/200 [08:41<14:24, 6.92s/it]
|
| 153 |
|
| 154 |
+
|
| 155 |
38%|ββββ | 75/200 [08:41<14:24, 6.92s/it]
|
| 156 |
38%|ββββ | 76/200 [08:48<14:16, 6.91s/it]
|
| 157 |
38%|ββββ | 77/200 [08:55<14:10, 6.92s/it]
|
| 158 |
39%|ββββ | 78/200 [09:02<13:58, 6.87s/it]
|
| 159 |
40%|ββββ | 79/200 [09:09<13:51, 6.88s/it]
|
| 160 |
40%|ββββ | 80/200 [09:15<13:43, 6.86s/it]
|
| 161 |
|
| 162 |
+
|
| 163 |
40%|ββββ | 80/200 [09:15<13:43, 6.86s/it]
|
| 164 |
40%|ββββ | 81/200 [09:22<13:36, 6.86s/it]
|
| 165 |
41%|ββββ | 82/200 [09:29<13:32, 6.89s/it]
|
| 166 |
42%|βββββ | 83/200 [09:36<13:22, 6.86s/it]
|
| 167 |
42%|βββββ | 84/200 [09:43<13:11, 6.82s/it]
|
| 168 |
42%|βββββ | 85/200 [09:50<13:09, 6.87s/it]
|
| 169 |
|
| 170 |
+
|
| 171 |
42%|βββββ | 85/200 [09:50<13:09, 6.87s/it]
|
| 172 |
43%|βββββ | 86/200 [09:57<13:03, 6.87s/it]
|
| 173 |
44%|βββββ | 87/200 [10:03<12:56, 6.87s/it]
|
| 174 |
44%|βββββ | 88/200 [10:10<12:49, 6.87s/it]
|
| 175 |
44%|βββββ | 89/200 [10:17<12:37, 6.83s/it]
|
| 176 |
45%|βββββ | 90/200 [10:24<12:31, 6.83s/it]
|
| 177 |
|
| 178 |
+
|
| 179 |
45%|βββββ | 90/200 [10:24<12:31, 6.83s/it]
|
| 180 |
46%|βββββ | 91/200 [10:31<12:27, 6.86s/it]
|
| 181 |
46%|βββββ | 92/200 [10:38<12:22, 6.87s/it]
|
| 182 |
46%|βββββ | 93/200 [10:45<12:12, 6.85s/it]
|
| 183 |
47%|βββββ | 94/200 [10:51<12:03, 6.82s/it]
|
| 184 |
48%|βββββ | 95/200 [10:58<11:56, 6.82s/it]
|
| 185 |
|
| 186 |
48%|βββββ | 95/200 [10:58<11:56, 6.82s/it]{'loss': '0.03869', 'grad_norm': '0.01334', 'learning_rate': '0.0001181', 'epoch': '3.393'}
|
| 187 |
+
|
| 188 |
48%|βββββ | 96/200 [11:05<11:52, 6.85s/it]
|
| 189 |
48%|βββββ | 97/200 [11:12<11:50, 6.90s/it]
|
| 190 |
49%|βββββ | 98/200 [11:19<11:45, 6.91s/it]
|
| 191 |
50%|βββββ | 99/200 [11:26<11:36, 6.89s/it]
|
| 192 |
50%|βββββ | 100/200 [11:33<11:27, 6.87s/it]
|
| 193 |
|
| 194 |
+
|
| 195 |
50%|βββββ | 100/200 [11:33<11:27, 6.87s/it]Unsloth: Restored added_tokens_decoder metadata in outputs/sft_deepseek_14b/checkpoint-100/tokenizer_config.json.
|
| 196 |
+
|
| 197 |
50%|βββββ | 101/200 [11:41<12:11, 7.39s/it]
|
| 198 |
51%|βββββ | 102/200 [11:48<11:49, 7.24s/it]
|
| 199 |
52%|ββββββ | 103/200 [11:55<11:30, 7.12s/it]
|
| 200 |
52%|ββββββ | 104/200 [12:02<11:15, 7.03s/it]
|
| 201 |
52%|ββββββ | 105/200 [12:09<11:03, 6.99s/it]
|
| 202 |
|
| 203 |
+
|
| 204 |
52%|ββββββ | 105/200 [12:09<11:03, 6.99s/it]
|
| 205 |
53%|ββββββ | 106/200 [12:16<10:53, 6.95s/it]
|
| 206 |
54%|ββββββ | 107/200 [12:22<10:42, 6.91s/it]
|
| 207 |
54%|ββββββ | 108/200 [12:29<10:31, 6.86s/it]
|
| 208 |
55%|ββββββ | 109/200 [12:36<10:20, 6.82s/it]
|
| 209 |
55%|ββββββ | 110/200 [12:43<10:12, 6.80s/it]
|
| 210 |
|
| 211 |
+
|
| 212 |
55%|ββββββ | 110/200 [12:43<10:12, 6.80s/it]
|
| 213 |
56%|ββββββ | 111/200 [12:49<10:07, 6.82s/it]
|
| 214 |
56%|ββββββ | 112/200 [12:56<10:04, 6.86s/it]
|
| 215 |
56%|ββββββ | 113/200 [13:04<10:09, 7.00s/it]
|
| 216 |
57%|ββββββ | 114/200 [13:11<09:59, 6.97s/it]
|
| 217 |
57%|ββββββ | 115/200 [13:18<09:51, 6.96s/it]
|
| 218 |
|
| 219 |
+
|
| 220 |
57%|ββββββ | 115/200 [13:18<09:51, 6.96s/it]
|
| 221 |
58%|ββββββ | 116/200 [13:25<09:43, 6.95s/it]
|
| 222 |
58%|ββββββ | 117/200 [13:31<09:31, 6.88s/it]
|
| 223 |
59%|ββββββ | 118/200 [13:38<09:24, 6.88s/it]
|
| 224 |
60%|ββββββ | 119/200 [13:45<09:18, 6.90s/it]
|
| 225 |
60%|ββββββ | 120/200 [13:52<09:12, 6.91s/it]
|
| 226 |
|
| 227 |
60%|ββββββ | 120/200 [13:52<09:12, 6.91s/it]{'loss': '0.03772', 'grad_norm': '0.01068', 'learning_rate': '7.706e-05', 'epoch': '4.286'}
|
| 228 |
+
|
| 229 |
60%|ββββββ | 121/200 [13:59<09:00, 6.84s/it]
|
| 230 |
61%|ββββββ | 122/200 [14:05<08:52, 6.83s/it]
|
| 231 |
62%|βββββββ | 123/200 [14:12<08:45, 6.82s/it]
|
| 232 |
62%|βββββββ | 124/200 [14:19<08:36, 6.80s/it]
|
| 233 |
62%|βββββββ | 125/200 [14:26<08:31, 6.83s/it]
|
| 234 |
|
| 235 |
62%|βββββββ | 125/200 [14:26<08:31, 6.83s/it]{'loss': '0.039', 'grad_norm': '0.01277', 'learning_rate': '6.91e-05', 'epoch': '4.464'}
|
| 236 |
+
|
| 237 |
63%|βββββββ | 126/200 [14:33<08:25, 6.83s/it]
|
| 238 |
64%|βββββββ | 127/200 [14:40<08:18, 6.83s/it]
|
| 239 |
64%|βββββββ | 128/200 [14:46<08:09, 6.79s/it]
|
| 240 |
64%|βββββββ | 129/200 [14:53<08:04, 6.82s/it]
|
| 241 |
65%|βββββββ | 130/200 [15:00<08:00, 6.86s/it]
|
| 242 |
|
| 243 |
+
|
| 244 |
65%|βββββββ | 130/200 [15:00<08:00, 6.86s/it]
|
| 245 |
66%|βββββββ | 131/200 [15:07<07:53, 6.87s/it]
|
| 246 |
66%|βββββββ | 132/200 [15:14<07:48, 6.89s/it]
|
| 247 |
66%|βββββββ | 133/200 [15:21<07:38, 6.84s/it]
|
| 248 |
67%|βββββββ | 134/200 [15:27<07:28, 6.80s/it]
|
| 249 |
68%|βββββββ | 135/200 [15:34<07:23, 6.82s/it]
|
| 250 |
|
| 251 |
68%|βββββββ | 135/200 [15:34<07:23, 6.82s/it]{'loss': '0.03844', 'grad_norm': '0.01827', 'learning_rate': '5.387e-05', 'epoch': '4.821'}
|
| 252 |
+
|
| 253 |
68%|βββββββ | 136/200 [15:41<07:19, 6.87s/it]
|
| 254 |
68%|βββββββ | 137/200 [15:48<07:14, 6.90s/it]
|
| 255 |
69%|βββββββ | 138/200 [15:55<07:07, 6.89s/it]
|
| 256 |
70%|βββββββ | 139/200 [16:02<06:58, 6.86s/it]
|
| 257 |
70%|βββββββ | 140/200 [16:09<06:52, 6.88s/it]
|
| 258 |
|
| 259 |
70%|βββββββ | 140/200 [16:09<06:52, 6.88s/it]{'loss': '0.03757', 'grad_norm': '0.01235', 'learning_rate': '4.67e-05', 'epoch': '5'}
|
| 260 |
+
|
| 261 |
70%|βββββββ | 141/200 [16:16<06:44, 6.85s/it]
|
| 262 |
71%|βββββββ | 142/200 [16:23<06:39, 6.90s/it]
|
| 263 |
72%|ββββββββ | 143/200 [16:29<06:31, 6.86s/it]
|
| 264 |
72%|ββββββββ | 144/200 [16:36<06:23, 6.85s/it]
|
| 265 |
72%|ββββββββ | 145/200 [16:43<06:17, 6.87s/it]
|
| 266 |
|
| 267 |
+
|
| 268 |
72%|ββββββββ | 145/200 [16:43<06:17, 6.87s/it]
|
| 269 |
73%|ββββββββ | 146/200 [16:50<06:11, 6.88s/it]
|
| 270 |
74%|ββββββββ | 147/200 [16:57<06:06, 6.91s/it]
|
| 271 |
74%|ββββββββ | 148/200 [17:04<06:00, 6.93s/it]
|
| 272 |
74%|ββββββββ | 149/200 [17:11<05:52, 6.91s/it]
|
| 273 |
75%|ββββββββ | 150/200 [17:18<05:44, 6.88s/it]
|
| 274 |
|
| 275 |
+
|
| 276 |
75%|ββββββββ | 150/200 [17:18<05:44, 6.88s/it]Unsloth: Restored added_tokens_decoder metadata in outputs/sft_deepseek_14b/checkpoint-150/tokenizer_config.json.
|
| 277 |
+
|
| 278 |
76%|ββββββββ | 151/200 [17:26<06:02, 7.41s/it]
|
| 279 |
76%|ββββββββ | 152/200 [17:33<05:47, 7.23s/it]
|
| 280 |
76%|ββββββββ | 153/200 [17:40<05:34, 7.11s/it]
|
| 281 |
77%|ββββββββ | 154/200 [17:47<05:26, 7.09s/it]
|
| 282 |
78%|ββββββββ | 155/200 [17:54<05:16, 7.03s/it]
|
| 283 |
|
| 284 |
+
|
| 285 |
78%|ββββββββ | 155/200 [17:54<05:16, 7.03s/it]
|
| 286 |
78%|ββββββββ | 156/200 [18:01<05:06, 6.97s/it]
|
| 287 |
78%|ββββββββ | 157/200 [18:07<04:57, 6.91s/it]
|
| 288 |
79%|ββββββββ | 158/200 [18:14<04:48, 6.87s/it]
|
| 289 |
80%|ββββββββ | 159/200 [18:21<04:41, 6.88s/it]
|
| 290 |
80%|ββββββββ | 160/200 [18:28<04:34, 6.87s/it]
|
| 291 |
|
| 292 |
+
|
| 293 |
80%|ββββββββ | 160/200 [18:28<04:34, 6.87s/it]
|
| 294 |
80%|ββββββοΏ½οΏ½β | 161/200 [18:35<04:26, 6.83s/it]
|
| 295 |
81%|ββββββββ | 162/200 [18:41<04:18, 6.81s/it]
|
| 296 |
82%|βββββββββ | 163/200 [18:48<04:12, 6.81s/it]
|
| 297 |
82%|βββββββββ | 164/200 [18:55<04:05, 6.83s/it]
|
| 298 |
82%|βββββββββ | 165/200 [19:02<03:59, 6.84s/it]
|
| 299 |
|
| 300 |
+
|
| 301 |
82%|βββββββββ | 165/200 [19:02<03:59, 6.84s/it]
|
| 302 |
83%|βββββββββ | 166/200 [19:09<03:53, 6.86s/it]
|
| 303 |
84%|βββββββββ | 167/200 [19:16<03:46, 6.87s/it]
|
| 304 |
84%|βββββββββ | 168/200 [19:23<03:38, 6.82s/it]
|
| 305 |
84%|βββββββββ | 169/200 [19:29<03:31, 6.83s/it]
|
| 306 |
85%|βββββββββ | 170/200 [19:36<03:24, 6.82s/it]
|
| 307 |
|
| 308 |
85%|βββββββββ | 170/200 [19:36<03:24, 6.82s/it]{'loss': '0.03837', 'grad_norm': '0.01728', 'learning_rate': '1.285e-05', 'epoch': '6.071'}
|
| 309 |
+
|
| 310 |
86%|βββββββββ | 171/200 [19:43<03:18, 6.86s/it]
|
| 311 |
86%|βββββββββ | 172/200 [19:50<03:11, 6.86s/it]
|
| 312 |
86%|βββββββββ | 173/200 [19:57<03:04, 6.84s/it]
|
| 313 |
87%|βββββββββ | 174/200 [20:04<02:58, 6.87s/it]
|
| 314 |
88%|βββββββββ | 175/200 [20:11<02:51, 6.86s/it]
|
| 315 |
|
| 316 |
+
|
| 317 |
88%|βββββββββ | 175/200 [20:11<02:51, 6.86s/it]
|
| 318 |
88%|βββββββββ | 176/200 [20:17<02:44, 6.86s/it]
|
| 319 |
88%|βββββββββ | 177/200 [20:24<02:38, 6.91s/it]
|
| 320 |
89%|βββββββββ | 178/200 [20:31<02:32, 6.91s/it]
|
| 321 |
90%|βββββββββ | 179/200 [20:38<02:25, 6.92s/it]
|
| 322 |
90%|βββββββββ | 180/200 [20:45<02:17, 6.87s/it]
|
| 323 |
|
| 324 |
90%|βββββββββ | 180/200 [20:45<02:17, 6.87s/it]{'loss': '0.03702', 'grad_norm': '0.01512', 'learning_rate': '5.968e-06', 'epoch': '6.429'}
|
| 325 |
+
|
| 326 |
90%|βββββββββ | 181/200 [20:52<02:11, 6.90s/it]
|
| 327 |
91%|βββββββββ | 182/200 [20:59<02:03, 6.85s/it]
|
| 328 |
92%|ββββββββββ| 183/200 [21:06<01:56, 6.84s/it]
|
| 329 |
92%|ββββββββββ| 184/200 [21:13<01:49, 6.86s/it]
|
| 330 |
92%|ββββββββββ| 185/200 [21:19<01:43, 6.88s/it]
|
| 331 |
|
| 332 |
+
|
| 333 |
92%|ββββββββββ| 185/200 [21:19<01:43, 6.88s/it]
|
| 334 |
93%|ββββββββββ| 186/200 [21:26<01:36, 6.88s/it]
|
| 335 |
94%|ββββββββββ| 187/200 [21:33<01:29, 6.91s/it]
|
| 336 |
94%|ββββββββββ| 188/200 [21:40<01:22, 6.83s/it]
|
| 337 |
94%|ββββββββββ| 189/200 [21:47<01:15, 6.85s/it]
|
| 338 |
95%|ββββββββββ| 190/200 [21:54<01:08, 6.85s/it]
|
| 339 |
|
| 340 |
95%|ββββββββββ| 190/200 [21:54<01:08, 6.85s/it]{'loss': '0.03777', 'grad_norm': '0.01418', 'learning_rate': '1.649e-06', 'epoch': '6.786'}
|
| 341 |
+
|
| 342 |
96%|ββββββββββ| 191/200 [22:01<01:01, 6.87s/it]
|
| 343 |
96%|ββββββββββ| 192/200 [22:07<00:54, 6.87s/it]
|
| 344 |
96%|ββββββββββ| 193/200 [22:14<00:47, 6.84s/it]
|
| 345 |
97%|ββββββββββ| 194/200 [22:21<00:40, 6.78s/it]
|
| 346 |
98%|ββββββββββ| 195/200 [22:28<00:33, 6.78s/it]
|
| 347 |
|
| 348 |
+
|
| 349 |
98%|ββββββββββ| 195/200 [22:28<00:33, 6.78s/it]
|
| 350 |
98%|ββββββββββ| 196/200 [22:35<00:27, 6.82s/it]
|
| 351 |
98%|ββββββββββ| 197/200 [22:42<00:20, 6.86s/it]
|
| 352 |
99%|ββββββββββ| 198/200 [22:48<00:13, 6.88s/it]
|
| 353 |
|
| 354 |
+
Unsloth: Restored added_tokens_decoder metadata in outputs/sft_deepseek_14b/checkpoint-200/tokenizer_config.json.
|
| 355 |
+
|
| 356 |
|
| 357 |
+
|
| 358 |
+
|
| 359 |
+
|
| 360 |
+
|
| 361 |
+
|
| 362 |
...eek_14b/training_args.bin: 100%|ββββββββββ| 5.78kB / 5.78kB [A[A
|
| 363 |
+
|
| 364 |
+
|
| 365 |
+
|
| 366 |
...epseek_14b/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB [A[A[A
|
| 367 |
+
|
| 368 |
+
|
| 369 |
+
|
| 370 |
+
|
| 371 |
...adapter_model.safetensors: 90%|βββββββββ | 248MB / 275MB [A[A[A[A
|
| 372 |
+
|
| 373 |
+
|
| 374 |
...eek_14b/training_args.bin: 100%|ββββββββββ| 5.78kB / 5.78kB [A[A
|
| 375 |
+
|
| 376 |
+
|
| 377 |
+
|
| 378 |
...epseek_14b/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB [A[A[A
|
| 379 |
+
|
| 380 |
+
|
| 381 |
+
|
| 382 |
+
|
| 383 |
...adapter_model.safetensors: 90%|βββββββββ | 248MB / 275MB [A[A[A[A
|
| 384 |
+
|
| 385 |
+
|
| 386 |
...eek_14b/training_args.bin: 100%|ββββββββββ| 5.78kB / 5.78kB [A[A
|
| 387 |
+
|
| 388 |
+
|
| 389 |
+
|
| 390 |
...epseek_14b/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB [A[A[A
|
| 391 |
+
|
| 392 |
+
|
| 393 |
+
|
| 394 |
+
|
| 395 |
...adapter_model.safetensors: 100%|ββββββββββ| 275MB / 275MB [A[A[A[A
|
| 396 |
+
|
| 397 |
+
|
| 398 |
...eek_14b/training_args.bin: 100%|ββββββββββ| 5.78kB / 5.78kB [A[A
|
| 399 |
+
|
| 400 |
+
|
| 401 |
+
|
| 402 |
...epseek_14b/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB [A[A[A
|
| 403 |
+
|
| 404 |
+
|
| 405 |
+
|
| 406 |
+
|
| 407 |
...adapter_model.safetensors: 100%|ββββββββββ| 275MB / 275MB [A[A[A[A
|
| 408 |
+
|
| 409 |
+
|
| 410 |
...eek_14b/training_args.bin: 100%|ββββββββββ| 5.78kB / 5.78kB [A[A
|
| 411 |
+
|
| 412 |
+
|
| 413 |
+
|
| 414 |
...epseek_14b/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB [A[A[A
|
| 415 |
+
|
| 416 |
+
|
| 417 |
+
|
| 418 |
+
|
| 419 |
...adapter_model.safetensors: 100%|ββββββββββ| 275MB / 275MB [A[A[A[A
|
| 420 |
+
|
| 421 |
+
|
| 422 |
...eek_14b/training_args.bin: 100%|ββββββββββ| 5.78kB / 5.78kB [A[A
|
| 423 |
+
|
| 424 |
+
|
| 425 |
+
|
| 426 |
...epseek_14b/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB [A[A[A
|
| 427 |
+
|
| 428 |
+
|
| 429 |
+
|
| 430 |
+
|
| 431 |
...adapter_model.safetensors: 100%|ββββββββββ| 275MB / 275MB [A[A[A[A
|
| 432 |
+
|
| 433 |
+
|
| 434 |
...eek_14b/training_args.bin: 100%|ββββββββββ| 5.78kB / 5.78kB
|
| 435 |
+
|
| 436 |
...epseek_14b/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB
|
| 437 |
+
|
| 438 |
...adapter_model.safetensors: 100%|ββββββββββ| 275MB / 275MB
|
| 439 |
+
|
| 440 |
+
|
| 441 |
+
|
| 442 |
+
|
| 443 |
...eek_14b/training_args.bin: 100%|ββββββββββ| 5.78kB / 5.78kB [A[A
|
| 444 |
+
|
| 445 |
+
|
| 446 |
+
|
| 447 |
...epseek_14b/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB [A[A[A
|
| 448 |
+
|
| 449 |
+
|
| 450 |
+
|
| 451 |
+
|
| 452 |
...adapter_model.safetensors: 96%|ββββββββββ| 264MB / 275MB [A[A[A[A
|
| 453 |
+
|
| 454 |
+
|
| 455 |
...eek_14b/training_args.bin: 100%|ββββββββββ| 5.78kB / 5.78kB [A[A
|
| 456 |
+
|
| 457 |
+
|
| 458 |
+
|
| 459 |
...epseek_14b/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB [A[A[A
|
| 460 |
+
|
| 461 |
+
|
| 462 |
+
|
| 463 |
+
|
| 464 |
...adapter_model.safetensors: 96%|ββββββββββ| 264MB / 275MB [A[A[A[A
|
| 465 |
+
|
| 466 |
+
|
| 467 |
...eek_14b/training_args.bin: 100%|ββββββββββ| 5.78kB / 5.78kB [A[A
|
| 468 |
+
|
| 469 |
+
|
| 470 |
+
|
| 471 |
...epseek_14b/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB [A[A[A
|
| 472 |
+
|
| 473 |
+
|
| 474 |
+
|
| 475 |
+
|
| 476 |
...adapter_model.safetensors: 100%|ββββββββββ| 275MB / 275MB [A[A[A[A
|
| 477 |
+
|
| 478 |
+
|
| 479 |
...eek_14b/training_args.bin: 100%|ββββββββββ| 5.78kB / 5.78kB [A[A
|
| 480 |
+
|
| 481 |
+
|
| 482 |
+
|
| 483 |
...epseek_14b/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB [A[A[A
|
| 484 |
+
|
| 485 |
+
|
| 486 |
+
|
| 487 |
+
|
| 488 |
...adapter_model.safetensors: 100%|ββββββββββ| 275MB / 275MB [A[A[A[A
|
| 489 |
+
|
| 490 |
+
|
| 491 |
...eek_14b/training_args.bin: 100%|ββββββββββ| 5.78kB / 5.78kB [A[A
|
| 492 |
+
|
| 493 |
+
|
| 494 |
+
|
| 495 |
...epseek_14b/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB [A[A[A
|
| 496 |
+
|
| 497 |
+
|
| 498 |
+
|
| 499 |
+
|
| 500 |
...adapter_model.safetensors: 100%|ββββββββββ| 275MB / 275MB [A[A[A[A
|
| 501 |
+
|
| 502 |
+
|
| 503 |
...eek_14b/training_args.bin: 100%|ββββββββββ| 5.78kB / 5.78kB [A[A
|
| 504 |
+
|
| 505 |
+
|
| 506 |
+
|
| 507 |
...epseek_14b/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB [A[A[A
|
| 508 |
+
|
| 509 |
+
|
| 510 |
+
|
| 511 |
+
|
| 512 |
...adapter_model.safetensors: 100%|ββββββββββ| 275MB / 275MB [A[A[A[A
|
| 513 |
+
|
| 514 |
+
|
| 515 |
...eek_14b/training_args.bin: 100%|ββββββββββ| 5.78kB / 5.78kB
|
| 516 |
+
|
| 517 |
...epseek_14b/tokenizer.json: 100%|ββββββββββ| 11.4MB / 11.4MB
|
| 518 |
+
|
| 519 |
...adapter_model.safetensors: 100%|ββββββββββ| 275MB / 275MB
|
| 520 |
+
No files have been modified since last commit. Skipping to prevent empty commit.
|
| 521 |
+
[huggingface_hub.hf_api|WARNING]No files have been modified since last commit. Skipping to prevent empty commit.
|
| 522 |
+
Pushed SFT adapter: Navigam/deepseek-r1-distill-qwen-14b_sft
|
| 523 |
+
|
| 524 |
+
---
|
| 525 |
+
exit_code: 0
|
| 526 |
+
elapsed_ms: 1453390
|
| 527 |
+
ended_at: 2026-04-26T03:21:21.768Z
|
| 528 |
+
---
|
training_logs/546061.txt
ADDED
|
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
pid: 117327
|
| 3 |
+
cwd: "/teamspace/studios/this_studio/corp_gym"
|
| 4 |
+
command: "stdbuf -oL -eL python -u eval.py --policy hf --label sft --model Qwen/Qwen2.5-7B-Instruct --adapter outputs/sft_adapter --output results/sft_eval.jsonl 2>&1 | tee results/runs/sft_eval.log"
|
| 5 |
+
started_at: 2026-04-25T16:37:34.195Z
|
| 6 |
+
running_for_ms: 80007
|
| 7 |
+
---
|
| 8 |
+
Skipping import of cpp extensions due to incompatible torch version. Please upgrade to torch >= 2.11.0 (found 2.7.1+cu128).
|
| 9 |
+
`torch_dtype` is deprecated! Use `dtype` instead!
|
| 10 |
+
|
| 11 |
+
The following generation flags are not valid and may be ignored: ['temperature', 'top_p', 'top_k']. Set `TRANSFORMERS_VERBOSITY=info` for more details.
|
| 12 |
+
|
| 13 |
+
[sft] 3 episodes
|
| 14 |
+
- e1_launch_readiness: terminal_reward=0.910 pass_rate=1.000 success=1.000
|
| 15 |
+
- m1_budget_reallocation: terminal_reward=0.943 pass_rate=1.000 success=1.000
|
| 16 |
+
- h1_acquisition_defence: terminal_reward=0.604 pass_rate=0.583 success=0.000
|
| 17 |
+
|
| 18 |
+
Wrote results/sft_eval.jsonl
|
| 19 |
+
Wrote results/metadata.json
|
| 20 |
+
|
| 21 |
+
---
|
| 22 |
+
exit_code: 0
|
| 23 |
+
elapsed_ms: 80810
|
| 24 |
+
ended_at: 2026-04-25T16:38:55.005Z
|
| 25 |
+
---
|
training_logs/829019.txt
ADDED
|
@@ -0,0 +1,105 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 0 |
0%| | 0/8 [00:00<?, ?it/s]`use_return_dict` is deprecated! Use `return_dict` instead!
|
|
|
|
|
|
|
| 1 |
12%|ββ | 1/8 [00:07<00:54, 7.72s/it]
|
| 2 |
25%|βββ | 2/8 [00:11<00:30, 5.11s/it]
|
| 3 |
38%|ββββ | 3/8 [00:18<00:31, 6.29s/it]
|
| 4 |
50%|βββββ | 4/8 [00:22<00:21, 5.48s/it]
|
| 5 |
62%|βββββββ | 5/8 [00:31<00:20, 6.68s/it]
|
| 6 |
|
|
|
|
| 7 |
62%|βββββββ | 5/8 [00:31<00:20, 6.68s/it]
|
| 8 |
75%|ββββββββ | 6/8 [00:36<00:11, 5.95s/it]
|
| 9 |
88%|βββββββββ | 7/8 [00:45<00:06, 6.88s/it]
|
|
|
|
| 10 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 11 |
0%| | 0/8 [00:00<?, ?it/s]Unsloth: Will smartly offload gradients to save VRAM!
|
|
|
|
| 12 |
12%|ββ | 1/8 [00:06<00:42, 6.06s/it]
|
| 13 |
25%|βββ | 2/8 [00:08<00:24, 4.16s/it]
|
| 14 |
38%|ββββ | 3/8 [00:15<00:26, 5.24s/it]
|
| 15 |
50%|βββββ | 4/8 [00:18<00:17, 4.43s/it]
|
| 16 |
62%|βββββββ | 5/8 [00:24<00:15, 5.12s/it]
|
| 17 |
|
|
|
|
| 18 |
62%|βββββββ | 5/8 [00:24<00:15, 5.12s/it]
|
| 19 |
75%|ββββββββ | 6/8 [00:28<00:09, 4.55s/it]
|
| 20 |
88%|βββββββββ | 7/8 [00:36<00:05, 5.61s/it]
|
|
|
|
| 21 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
pid: 174904
|
| 3 |
+
cwd: "/teamspace/studios/this_studio/corp_gym"
|
| 4 |
+
command: "mkdir -p results/runs outputs/rlvr_verify\nexport HF_HUB_ENABLE_HF_TRANSFER=1\nexport CORP_STUB_WORKERS=1\nexport CORP_DISABLE_LLM_JUDGE=1\nexport TOKENIZERS_PARALLELISM=false\nexport TRANSFORMERS_VERBOSITY=warning\n\npython -u training/train_rlvr.py \\\n --model Qwen/Qwen2.5-7B-Instruct \\\n --adapter outputs/sft_adapter \\\n --examples data/processed/e1_m1_clean.jsonl,data/processed/h1_seed_clean.jsonl \\\n --output outputs/rlvr_verify \\\n --rounds 2 \\\n --n-samples 4 \\\n --max-prompts 12 \\\n --max-prompt-length 6144 \\\n --max-completion-length 384 \\\n --inner-lr 1e-4 \\\n --inner-epochs 1 \\\n --inner-max-steps 8 \\\n --stats-file results/runs/rlvr_verify_stats.jsonl \\\n 2>&1 | tee results/runs/rlvr_verify.log"
|
| 5 |
+
started_at: 2026-04-25T18:29:18.454Z
|
| 6 |
+
running_for_ms: 215008
|
| 7 |
+
---
|
| 8 |
+
π¦₯ Unsloth: Will patch your computer to enable 2x faster free finetuning.
|
| 9 |
+
π¦₯ Unsloth Zoo will now patch everything to make training faster!
|
| 10 |
+
Built 1384 prompts from data/processed/e1_m1_clean.jsonl,data/processed/h1_seed_clean.jsonl
|
| 11 |
+
==((====))== Unsloth 2026.4.8: Fast Qwen2 patching. Transformers: 5.5.0.
|
| 12 |
+
\\ /| NVIDIA H100 80GB HBM3. Num GPUs = 1. Max memory: 79.189 GB. Platform: Linux.
|
| 13 |
+
O^O/ \_/ \ Torch: 2.7.1+cu128. CUDA: 9.0. CUDA Toolkit: 12.8. Triton: 3.3.1
|
| 14 |
+
\ / Bfloat16 = TRUE. FA [Xformers = 0.0.31.post1. FA2 = True]
|
| 15 |
+
"-____-" Free license: http://github.com/unslothai/unsloth
|
| 16 |
+
Unsloth: Fast downloading is enabled - ignore downloading bars which are red colored!
|
| 17 |
+
|
| 18 |
+
unsloth/qwen2.5-7b-instruct-unsloth-bnb-4bit does not have a padding token! Will use pad_token = <|PAD_TOKEN|>.
|
| 19 |
+
Length filter: kept 996 prompts (<= 5529 tokens); skipped 388.
|
| 20 |
+
Loaded adapter: outputs/sft_adapter
|
| 21 |
+
|
| 22 |
+
=== RLVR round 1/2 ===
|
| 23 |
+
using 12 prompts
|
| 24 |
+
Both `max_new_tokens` (=384) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 25 |
+
/home/zeus/miniconda3/envs/cloudspace/lib/python3.12/site-packages/transformers/modeling_attn_mask_utils.py:71: FutureWarning: The attention mask API under `transformers.modeling_attn_mask_utils` (`AttentionMaskConverter`) is deprecated and will be removed in Transformers v5.10. Please use the new API in `transformers.masking_utils`.
|
| 26 |
+
warnings.warn(DEPRECATION_MESSAGE, FutureWarning)
|
| 27 |
+
/home/zeus/miniconda3/envs/cloudspace/lib/python3.12/site-packages/transformers/modeling_attn_mask_utils.py:281: FutureWarning: The attention mask API under `transformers.modeling_attn_mask_utils` (`AttentionMaskConverter`) is deprecated and will be removed in Transformers v5.10. Please use the new API in `transformers.masking_utils`.
|
| 28 |
+
warnings.warn(DEPRECATION_MESSAGE, FutureWarning)
|
| 29 |
+
Both `max_new_tokens` (=384) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 30 |
+
Both `max_new_tokens` (=384) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 31 |
+
Both `max_new_tokens` (=384) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 32 |
+
Both `max_new_tokens` (=384) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 33 |
+
Both `max_new_tokens` (=384) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 34 |
+
Both `max_new_tokens` (=384) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 35 |
+
Both `max_new_tokens` (=384) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 36 |
+
Both `max_new_tokens` (=384) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 37 |
+
Both `max_new_tokens` (=384) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 38 |
+
rollout 10/12 kept=10 mean_best=0.394 mean_any=0.394 elapsed=21.9s
|
| 39 |
+
Both `max_new_tokens` (=384) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 40 |
+
Both `max_new_tokens` (=384) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 41 |
+
rollout 12/12 kept=12 mean_best=0.360 mean_any=0.360 elapsed=26.2s
|
| 42 |
+
round 1 summary: kept=12/12 keep_rate=100.00% mean_best=0.360 mean_any=0.360 secs=26.2
|
| 43 |
+
warmup_ratio is deprecated and will be removed in v5.2. Use `warmup_steps` instead.
|
| 44 |
+
num_proc must be <= 12. Reducing num_proc to 12 for dataset of size 12.
|
| 45 |
+
[datasets.arrow_dataset|WARNING]num_proc must be <= 12. Reducing num_proc to 12 for dataset of size 12.
|
| 46 |
+
|
| 47 |
+
π¦₯ Unsloth: Padding-free auto-enabled, enabling faster training.
|
| 48 |
+
The tokenizer has new PAD/BOS/EOS tokens that differ from the model config and generation config. The model config and generation config were aligned accordingly, being updated with the tokenizer's values. Updated tokens: {'bos_token_id': None}.
|
| 49 |
+
==((====))== Unsloth - 2x faster free finetuning | Num GPUs used = 1
|
| 50 |
+
\\ /| Num examples = 12 | Num Epochs = 4 | Total steps = 8
|
| 51 |
+
O^O/ \_/ \ Batch size per device = 1 | Gradient accumulation steps = 8
|
| 52 |
+
\ / Data Parallel GPUs = 1 | Total batch size (1 x 8 x 1) = 8
|
| 53 |
+
"-____-" Trainable parameters = 80,740,352 of 7,696,356,864 (1.05% trained)
|
| 54 |
+
|
| 55 |
0%| | 0/8 [00:00<?, ?it/s]`use_return_dict` is deprecated! Use `return_dict` instead!
|
| 56 |
+
Unsloth: Will smartly offload gradients to save VRAM!
|
| 57 |
+
|
| 58 |
12%|ββ | 1/8 [00:07<00:54, 7.72s/it]
|
| 59 |
25%|βββ | 2/8 [00:11<00:30, 5.11s/it]
|
| 60 |
38%|ββββ | 3/8 [00:18<00:31, 6.29s/it]
|
| 61 |
50%|βββββ | 4/8 [00:22<00:21, 5.48s/it]
|
| 62 |
62%|βββββββ | 5/8 [00:31<00:20, 6.68s/it]
|
| 63 |
|
| 64 |
+
|
| 65 |
62%|βββββββ | 5/8 [00:31<00:20, 6.68s/it]
|
| 66 |
75%|ββββββββ | 6/8 [00:36<00:11, 5.95s/it]
|
| 67 |
88%|βββββββββ | 7/8 [00:45<00:06, 6.88s/it]
|
| 68 |
+
|
| 69 |
|
| 70 |
+
|
| 71 |
+
|
| 72 |
+
=== RLVR round 2/2 ===
|
| 73 |
+
using 12 prompts
|
| 74 |
+
Both `max_new_tokens` (=384) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 75 |
+
/home/zeus/miniconda3/envs/cloudspace/lib/python3.12/site-packages/transformers/modeling_attn_mask_utils.py:71: FutureWarning: The attention mask API under `transformers.modeling_attn_mask_utils` (`AttentionMaskConverter`) is deprecated and will be removed in Transformers v5.10. Please use the new API in `transformers.masking_utils`.
|
| 76 |
+
warnings.warn(DEPRECATION_MESSAGE, FutureWarning)
|
| 77 |
+
/home/zeus/miniconda3/envs/cloudspace/lib/python3.12/site-packages/transformers/modeling_attn_mask_utils.py:281: FutureWarning: The attention mask API under `transformers.modeling_attn_mask_utils` (`AttentionMaskConverter`) is deprecated and will be removed in Transformers v5.10. Please use the new API in `transformers.masking_utils`.
|
| 78 |
+
warnings.warn(DEPRECATION_MESSAGE, FutureWarning)
|
| 79 |
+
Both `max_new_tokens` (=384) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 80 |
+
Both `max_new_tokens` (=384) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 81 |
+
Both `max_new_tokens` (=384) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 82 |
+
Both `max_new_tokens` (=384) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 83 |
+
Both `max_new_tokens` (=384) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 84 |
+
Both `max_new_tokens` (=384) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 85 |
+
Both `max_new_tokens` (=384) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 86 |
+
Both `max_new_tokens` (=384) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 87 |
+
Both `max_new_tokens` (=384) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 88 |
+
rollout 10/12 kept=10 mean_best=0.242 mean_any=0.242 elapsed=28.8s
|
| 89 |
+
Both `max_new_tokens` (=384) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 90 |
+
Both `max_new_tokens` (=384) and `max_length`(=32768) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
|
| 91 |
+
rollout 12/12 kept=12 mean_best=0.225 mean_any=0.225 elapsed=34.8s
|
| 92 |
+
round 2 summary: kept=12/12 keep_rate=100.00% mean_best=0.225 mean_any=0.225 secs=34.8
|
| 93 |
+
warmup_ratio is deprecated and will be removed in v5.2. Use `warmup_steps` instead.
|
| 94 |
+
num_proc must be <= 12. Reducing num_proc to 12 for dataset of size 12.
|
| 95 |
+
[datasets.arrow_dataset|WARNING]num_proc must be <= 12. Reducing num_proc to 12 for dataset of size 12.
|
| 96 |
+
|
| 97 |
+
π¦₯ Unsloth: Padding-free auto-enabled, enabling faster training.
|
| 98 |
+
==((====))== Unsloth - 2x faster free finetuning | Num GPUs used = 1
|
| 99 |
+
\\ /| Num examples = 12 | Num Epochs = 4 | Total steps = 8
|
| 100 |
+
O^O/ \_/ \ Batch size per device = 1 | Gradient accumulation steps = 8
|
| 101 |
+
\ / Data Parallel GPUs = 1 | Total batch size (1 x 8 x 1) = 8
|
| 102 |
+
"-____-" Trainable parameters = 80,740,352 of 7,696,356,864 (1.05% trained)
|
| 103 |
+
|
| 104 |
0%| | 0/8 [00:00<?, ?it/s]Unsloth: Will smartly offload gradients to save VRAM!
|
| 105 |
+
|
| 106 |
12%|ββ | 1/8 [00:06<00:42, 6.06s/it]
|
| 107 |
25%|βββ | 2/8 [00:08<00:24, 4.16s/it]
|
| 108 |
38%|ββββ | 3/8 [00:15<00:26, 5.24s/it]
|
| 109 |
50%|βββββ | 4/8 [00:18<00:17, 4.43s/it]
|
| 110 |
62%|βββββββ | 5/8 [00:24<00:15, 5.12s/it]
|
| 111 |
|
| 112 |
+
|
| 113 |
62%|βββββββ | 5/8 [00:24<00:15, 5.12s/it]
|
| 114 |
75%|ββββββββ | 6/8 [00:28<00:09, 4.55s/it]
|
| 115 |
88%|βββββββββ | 7/8 [00:36<00:05, 5.61s/it]
|
| 116 |
+
|
| 117 |
|
| 118 |
+
|
| 119 |
+
Unsloth: Restored added_tokens_decoder metadata in outputs/rlvr_verify/tokenizer_config.json.
|
| 120 |
+
|
| 121 |
+
Saved RLVR adapter to outputs/rlvr_verify
|
| 122 |
+
|
| 123 |
+
---
|
| 124 |
+
exit_code: 0
|
| 125 |
+
elapsed_ms: 218590
|
| 126 |
+
ended_at: 2026-04-25T18:32:57.044Z
|
| 127 |
+
---
|