Training in progress, step 10
Browse files- adapter_config.json +3 -3
- adapter_model.safetensors +1 -1
- args.json +5 -5
- logging.jsonl +13 -18
- runs/events.out.tfevents.1743269896.notebook-67532d59-da58-11ef-a92b-d22a634ff48d-0.262569.0 +3 -0
- training_args.bin +1 -1
- val_dataset.jsonl +0 -0
adapter_config.json
CHANGED
|
@@ -24,12 +24,12 @@
|
|
| 24 |
"revision": null,
|
| 25 |
"target_modules": [
|
| 26 |
"v_proj",
|
| 27 |
-
"o_proj",
|
| 28 |
-
"down_proj",
|
| 29 |
"k_proj",
|
|
|
|
|
|
|
| 30 |
"gate_proj",
|
| 31 |
"q_proj",
|
| 32 |
-
"
|
| 33 |
],
|
| 34 |
"task_type": "CAUSAL_LM",
|
| 35 |
"use_dora": false,
|
|
|
|
| 24 |
"revision": null,
|
| 25 |
"target_modules": [
|
| 26 |
"v_proj",
|
|
|
|
|
|
|
| 27 |
"k_proj",
|
| 28 |
+
"up_proj",
|
| 29 |
+
"down_proj",
|
| 30 |
"gate_proj",
|
| 31 |
"q_proj",
|
| 32 |
+
"o_proj"
|
| 33 |
],
|
| 34 |
"task_type": "CAUSAL_LM",
|
| 35 |
"use_dora": false,
|
adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1478569432
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7800b88d8b3b3aff0dede77f3ad4f85ed2185b764d10ebbc00a745834d5c5500
|
| 3 |
size 1478569432
|
args.json
CHANGED
|
@@ -24,7 +24,7 @@
|
|
| 24 |
"use_chat_template": true,
|
| 25 |
"template_backend": "swift",
|
| 26 |
"dataset": [
|
| 27 |
-
"
|
| 28 |
],
|
| 29 |
"val_dataset": [],
|
| 30 |
"split_dataset_ratio": 0.03,
|
|
@@ -78,7 +78,7 @@
|
|
| 78 |
"custom_register_path": [],
|
| 79 |
"ignore_args_error": false,
|
| 80 |
"use_swift_lora": false,
|
| 81 |
-
"output_dir": "/root/dataDisk/ms-swift/output/
|
| 82 |
"overwrite_output_dir": false,
|
| 83 |
"do_train": false,
|
| 84 |
"do_eval": false,
|
|
@@ -108,7 +108,7 @@
|
|
| 108 |
"log_level": "passive",
|
| 109 |
"log_level_replica": "warning",
|
| 110 |
"log_on_each_node": true,
|
| 111 |
-
"logging_dir": "/root/dataDisk/ms-swift/output/
|
| 112 |
"logging_strategy": "steps",
|
| 113 |
"logging_first_step": true,
|
| 114 |
"logging_steps": 1,
|
|
@@ -351,8 +351,8 @@
|
|
| 351 |
"local_world_size": 8,
|
| 352 |
"model_suffix": "mistral-agentic-turbo",
|
| 353 |
"model_info": "ModelInfo(model_type='mistral_2501', model_dir='/root/highspeedstorage/ft-volume/mistral-agentic-turbo', torch_dtype=torch.bfloat16, max_model_len=131072, quant_method=None, quant_bits=None, rope_scaling={'factor': 4.0, 'original_max_position_embeddings': 32768, 'type': 'dynamic'}, config=None, task_type='causal_lm', num_labels=None)",
|
| 354 |
-
"model_meta": "ModelMeta(model_type='mistral_2501', model_groups=[ModelGroup(models=[Model(ms_model_id='mistralai/Mistral-Small-24B-Base-2501', hf_model_id='mistralai/Mistral-Small-24B-Base-2501', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='mistralai/Mistral-Small-24B-Instruct-2501', hf_model_id='mistralai/Mistral-Small-24B-Instruct-2501', model_path=None, ms_revision=None, hf_revision=None)], ignore_patterns=None, requires=None, tags=[])], template='mistral_2501', get_function=<function get_model_tokenizer_with_flash_attn at
|
| 355 |
"model_dir": "/root/highspeedstorage/ft-volume/mistral-agentic-turbo",
|
| 356 |
"hub": "<class 'swift.hub.hub.HFHub'>",
|
| 357 |
-
"training_args": "Seq2SeqTrainingArguments(output_dir='/root/dataDisk/ms-swift/output/
|
| 358 |
}
|
|
|
|
| 24 |
"use_chat_template": true,
|
| 25 |
"template_backend": "swift",
|
| 26 |
"dataset": [
|
| 27 |
+
"filtered_processed_logs.jsonl"
|
| 28 |
],
|
| 29 |
"val_dataset": [],
|
| 30 |
"split_dataset_ratio": 0.03,
|
|
|
|
| 78 |
"custom_register_path": [],
|
| 79 |
"ignore_args_error": false,
|
| 80 |
"use_swift_lora": false,
|
| 81 |
+
"output_dir": "/root/dataDisk/ms-swift/output/v39-20250329-173701",
|
| 82 |
"overwrite_output_dir": false,
|
| 83 |
"do_train": false,
|
| 84 |
"do_eval": false,
|
|
|
|
| 108 |
"log_level": "passive",
|
| 109 |
"log_level_replica": "warning",
|
| 110 |
"log_on_each_node": true,
|
| 111 |
+
"logging_dir": "/root/dataDisk/ms-swift/output/v39-20250329-173701/runs",
|
| 112 |
"logging_strategy": "steps",
|
| 113 |
"logging_first_step": true,
|
| 114 |
"logging_steps": 1,
|
|
|
|
| 351 |
"local_world_size": 8,
|
| 352 |
"model_suffix": "mistral-agentic-turbo",
|
| 353 |
"model_info": "ModelInfo(model_type='mistral_2501', model_dir='/root/highspeedstorage/ft-volume/mistral-agentic-turbo', torch_dtype=torch.bfloat16, max_model_len=131072, quant_method=None, quant_bits=None, rope_scaling={'factor': 4.0, 'original_max_position_embeddings': 32768, 'type': 'dynamic'}, config=None, task_type='causal_lm', num_labels=None)",
|
| 354 |
+
"model_meta": "ModelMeta(model_type='mistral_2501', model_groups=[ModelGroup(models=[Model(ms_model_id='mistralai/Mistral-Small-24B-Base-2501', hf_model_id='mistralai/Mistral-Small-24B-Base-2501', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='mistralai/Mistral-Small-24B-Instruct-2501', hf_model_id='mistralai/Mistral-Small-24B-Instruct-2501', model_path=None, ms_revision=None, hf_revision=None)], ignore_patterns=None, requires=None, tags=[])], template='mistral_2501', get_function=<function get_model_tokenizer_with_flash_attn at 0x7fdeb5812c20>, model_arch='llama', architectures=['MistralForCausalLM'], additional_saved_files=[], torch_dtype=None, is_multimodal=False, is_reward=False, task_type=None, ignore_patterns=[], requires=[], tags=[])",
|
| 355 |
"model_dir": "/root/highspeedstorage/ft-volume/mistral-agentic-turbo",
|
| 356 |
"hub": "<class 'swift.hub.hub.HFHub'>",
|
| 357 |
+
"training_args": "Seq2SeqTrainingArguments(output_dir='/root/dataDisk/ms-swift/output/v39-20250329-173701', overwrite_output_dir=False, do_train=False, do_eval=True, do_predict=False, eval_strategy=<IntervalStrategy.STEPS: 'steps'>, prediction_loss_only=False, per_device_train_batch_size=1, per_device_eval_batch_size=1, per_gpu_train_batch_size=None, per_gpu_eval_batch_size=None, gradient_accumulation_steps=8, eval_accumulation_steps=None, eval_delay=0, torch_empty_cache_steps=None, learning_rate=1e-05, weight_decay=0.01, adam_beta1=0.9, adam_beta2=0.999, adam_epsilon=1e-08, max_grad_norm=1.0, num_train_epochs=1.0, max_steps=-1, lr_scheduler_type=<SchedulerType.LINEAR: 'linear'>, lr_scheduler_kwargs=None, warmup_ratio=0.1, warmup_steps=0, log_level='passive', log_level_replica='warning', log_on_each_node=True, logging_dir='/root/dataDisk/ms-swift/output/v39-20250329-173701/runs', logging_strategy=<IntervalStrategy.STEPS: 'steps'>, logging_first_step=True, logging_steps=1, logging_nan_inf_filter=True, save_strategy=<SaveStrategy.STEPS: 'steps'>, save_steps=10, save_total_limit=2, save_safetensors=True, save_on_each_node=False, save_only_model=False, restore_callback_states_from_checkpoint=False, no_cuda=False, use_cpu=False, use_mps_device=False, seed=42, data_seed=42, jit_mode_eval=False, use_ipex=False, bf16=True, fp16=False, fp16_opt_level='O1', half_precision_backend='auto', bf16_full_eval=False, fp16_full_eval=False, tf32=None, local_rank=0, ddp_backend=None, tpu_num_cores=None, tpu_metrics_debug=False, debug=[], dataloader_drop_last=False, eval_steps=10, dataloader_num_workers=0, dataloader_prefetch_factor=None, past_index=-1, run_name='/root/dataDisk/ms-swift/output/v39-20250329-173701', disable_tqdm=False, remove_unused_columns=False, label_names=None, load_best_model_at_end=False, metric_for_best_model='loss', greater_is_better=False, ignore_data_skip=False, fsdp=[], fsdp_min_num_params=0, fsdp_config={'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, tp_size=0, fsdp_transformer_layer_cls_to_wrap=None, accelerator_config=AcceleratorConfig(split_batches=False, dispatch_batches=False, even_batches=True, use_seedable_sampler=True, non_blocking=False, gradient_accumulation_kwargs=None, use_configured_state=False), deepspeed={'fp16': {'enabled': 'auto', 'loss_scale': 0, 'loss_scale_window': 1000, 'initial_scale_power': 16, 'hysteresis': 2, 'min_loss_scale': 1}, 'bf16': {'enabled': 'auto'}, 'zero_optimization': {'stage': 3, 'offload_optimizer': {'device': 'cpu', 'pin_memory': True}, 'offload_param': {'device': 'cpu', 'pin_memory': True}, 'overlap_comm': True, 'contiguous_gradients': True, 'sub_group_size': 1000000000.0, 'reduce_bucket_size': 'auto', 'stage3_prefetch_bucket_size': 'auto', 'stage3_param_persistence_threshold': 'auto', 'stage3_max_live_parameters': 1000000000.0, 'stage3_max_reuse_distance': 1000000000.0, 'stage3_gather_16bit_weights_on_model_save': True}, 'gradient_accumulation_steps': 'auto', 'gradient_clipping': 'auto', 'steps_per_print': 2000, 'train_batch_size': 'auto', 'train_micro_batch_size_per_gpu': 'auto', 'wall_clock_breakdown': False}, label_smoothing_factor=0.0, optim=<OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, optim_args=None, adafactor=False, group_by_length=False, length_column_name='length', report_to=['tensorboard'], ddp_find_unused_parameters=None, ddp_bucket_cap_mb=None, ddp_broadcast_buffers=None, dataloader_pin_memory=True, dataloader_persistent_workers=False, skip_memory_metrics=True, use_legacy_prediction_loop=False, push_to_hub=True, resume_from_checkpoint=None, hub_model_id='TheAgenticAI/LLAMA-3.3-70B-Reasoning', hub_strategy=<HubStrategy.EVERY_SAVE: 'every_save'>, hub_token=None, hub_private_repo=True, hub_always_push=False, gradient_checkpointing=True, gradient_checkpointing_kwargs={'use_reentrant': True}, include_inputs_for_metrics=False, include_for_metrics=[], eval_do_concat_batches=True, fp16_backend='auto', evaluation_strategy='steps', push_to_hub_model_id=None, push_to_hub_organization=None, push_to_hub_token=None, mp_parameters='', auto_find_batch_size=False, full_determinism=False, torchdynamo=None, ray_scope='last', ddp_timeout=1800, torch_compile=False, torch_compile_backend=None, torch_compile_mode=None, dispatch_batches=None, split_batches=None, include_tokens_per_second=None, include_num_input_tokens_seen=None, neftune_noise_alpha=None, optim_target_modules=None, batch_eval_metrics=False, eval_on_start=False, use_liger_kernel=False, eval_use_gather_object=False, average_tokens_across_devices=None, sortish_sampler=False, predict_with_generate=False, generation_max_length=None, generation_num_beams=None, generation_config=None, acc_strategy='token', sequence_parallel_size=1, check_model=True, train_sampler_random=True, is_encoder_decoder=False, metric_warmup_step=0, train_dataset_sample=-1, fsdp_num=1, acc_steps=1, train_type='lora', optimizer=None, local_repo_path=None, galore_config=None)"
|
| 358 |
}
|
logging.jsonl
CHANGED
|
@@ -1,18 +1,13 @@
|
|
| 1 |
-
{"loss": 0.
|
| 2 |
-
{"loss": 0.
|
| 3 |
-
{"loss": 0.
|
| 4 |
-
{"loss": 0.
|
| 5 |
-
{"loss": 0.
|
| 6 |
-
{"loss": 0.
|
| 7 |
-
{"loss": 0.
|
| 8 |
-
{"loss": 0.
|
| 9 |
-
{"loss": 0.
|
| 10 |
-
{"loss": 0.
|
| 11 |
-
{"eval_loss": 0.
|
| 12 |
-
{"loss": 0.
|
| 13 |
-
{"loss": 0.
|
| 14 |
-
{"loss": 0.15041173, "token_acc": 0.94953342, "grad_norm": 4.02991343, "learning_rate": 1.54e-06, "memory(GiB)": 128.91, "train_speed(iter/s)": 0.019883, "epoch": 0.85245902, "global_step/max_steps": "13/15", "percentage": "86.67%", "elapsed_time": "10m 44s", "remaining_time": "1m 39s"}
|
| 15 |
-
{"loss": 0.1629689, "token_acc": 0.96349914, "grad_norm": 3.61858225, "learning_rate": 7.7e-07, "memory(GiB)": 128.91, "train_speed(iter/s)": 0.020077, "epoch": 0.91803279, "global_step/max_steps": "14/15", "percentage": "93.33%", "elapsed_time": "11m 28s", "remaining_time": "49s"}
|
| 16 |
-
{"loss": 0.11973689, "token_acc": 0.96463922, "grad_norm": 3.35976338, "learning_rate": 0.0, "memory(GiB)": 128.91, "train_speed(iter/s)": 0.01962, "epoch": 0.98360656, "global_step/max_steps": "15/15", "percentage": "100.00%", "elapsed_time": "12m 35s", "remaining_time": "0s"}
|
| 17 |
-
{"eval_loss": 0.15088241, "eval_token_acc": 0.96354056, "eval_runtime": 5.2152, "eval_samples_per_second": 5.752, "eval_steps_per_second": 0.767, "epoch": 0.98360656, "global_step/max_steps": "15/15", "percentage": "100.00%", "elapsed_time": "12m 40s", "remaining_time": "0s"}
|
| 18 |
-
{"train_runtime": 766.5904, "train_samples_per_second": 1.269, "train_steps_per_second": 0.02, "total_flos": 17121736916992.0, "train_loss": 0.26244897, "epoch": 0.98360656, "global_step/max_steps": "15/15", "percentage": "100.00%", "elapsed_time": "12m 46s", "remaining_time": "0s"}
|
|
|
|
| 1 |
+
{"loss": 0.4534499, "token_acc": 0.85207855, "grad_norm": 13.34440231, "learning_rate": 5e-06, "memory(GiB)": 59.77, "train_speed(iter/s)": 0.017298, "epoch": 0.0733945, "global_step/max_steps": "1/13", "percentage": "7.69%", "elapsed_time": "50s", "remaining_time": "10m 4s"}
|
| 2 |
+
{"loss": 0.48438707, "token_acc": 0.84307111, "grad_norm": 14.0229454, "learning_rate": 1e-05, "memory(GiB)": 97.94, "train_speed(iter/s)": 0.022243, "epoch": 0.14678899, "global_step/max_steps": "2/13", "percentage": "15.38%", "elapsed_time": "1m 22s", "remaining_time": "7m 33s"}
|
| 3 |
+
{"loss": 0.39558721, "token_acc": 0.87623521, "grad_norm": 17.53300667, "learning_rate": 9.09e-06, "memory(GiB)": 100.44, "train_speed(iter/s)": 0.023304, "epoch": 0.22018349, "global_step/max_steps": "3/13", "percentage": "23.08%", "elapsed_time": "2m 1s", "remaining_time": "6m 44s"}
|
| 4 |
+
{"loss": 0.36164382, "token_acc": 0.88360022, "grad_norm": 8.0753746, "learning_rate": 8.18e-06, "memory(GiB)": 100.44, "train_speed(iter/s)": 0.024548, "epoch": 0.29357798, "global_step/max_steps": "4/13", "percentage": "30.77%", "elapsed_time": "2m 35s", "remaining_time": "5m 49s"}
|
| 5 |
+
{"loss": 0.28607693, "token_acc": 0.89519958, "grad_norm": 4.6801033, "learning_rate": 7.27e-06, "memory(GiB)": 100.44, "train_speed(iter/s)": 0.02483, "epoch": 0.36697248, "global_step/max_steps": "5/13", "percentage": "38.46%", "elapsed_time": "3m 13s", "remaining_time": "5m 10s"}
|
| 6 |
+
{"loss": 0.24699311, "token_acc": 0.92354802, "grad_norm": 4.40268326, "learning_rate": 6.36e-06, "memory(GiB)": 118.38, "train_speed(iter/s)": 0.024078, "epoch": 0.44036697, "global_step/max_steps": "6/13", "percentage": "46.15%", "elapsed_time": "4m 1s", "remaining_time": "4m 42s"}
|
| 7 |
+
{"loss": 0.20653641, "token_acc": 0.92372492, "grad_norm": 4.65564156, "learning_rate": 5.45e-06, "memory(GiB)": 118.38, "train_speed(iter/s)": 0.024314, "epoch": 0.51376147, "global_step/max_steps": "7/13", "percentage": "53.85%", "elapsed_time": "4m 40s", "remaining_time": "4m 0s"}
|
| 8 |
+
{"loss": 0.17608178, "token_acc": 0.93256823, "grad_norm": 4.18540716, "learning_rate": 4.55e-06, "memory(GiB)": 118.38, "train_speed(iter/s)": 0.024257, "epoch": 0.58715596, "global_step/max_steps": "8/13", "percentage": "61.54%", "elapsed_time": "5m 22s", "remaining_time": "3m 21s"}
|
| 9 |
+
{"loss": 0.17281821, "token_acc": 0.94667173, "grad_norm": 4.15249395, "learning_rate": 3.64e-06, "memory(GiB)": 118.38, "train_speed(iter/s)": 0.024331, "epoch": 0.66055046, "global_step/max_steps": "9/13", "percentage": "69.23%", "elapsed_time": "6m 2s", "remaining_time": "2m 41s"}
|
| 10 |
+
{"loss": 0.1793212, "token_acc": 0.96042697, "grad_norm": 4.5847702, "learning_rate": 2.73e-06, "memory(GiB)": 131.34, "train_speed(iter/s)": 0.023399, "epoch": 0.73394495, "global_step/max_steps": "10/13", "percentage": "76.92%", "elapsed_time": "6m 59s", "remaining_time": "2m 5s"}
|
| 11 |
+
{"eval_loss": 0.17151441, "eval_token_acc": 0.94895582, "eval_runtime": 5.227, "eval_samples_per_second": 4.974, "eval_steps_per_second": 0.765, "epoch": 0.73394495, "global_step/max_steps": "10/13", "percentage": "76.92%", "elapsed_time": "7m 5s", "remaining_time": "2m 7s"}
|
| 12 |
+
{"loss": 0.15501738, "token_acc": 0.9503762, "grad_norm": 4.98172522, "learning_rate": 1.82e-06, "memory(GiB)": 131.34, "train_speed(iter/s)": 0.022982, "epoch": 0.80733945, "global_step/max_steps": "11/13", "percentage": "84.62%", "elapsed_time": "7m 51s", "remaining_time": "1m 25s"}
|
| 13 |
+
{"loss": 0.14941445, "token_acc": 0.94386522, "grad_norm": 3.96140695, "learning_rate": 9.1e-07, "memory(GiB)": 131.34, "train_speed(iter/s)": 0.023428, "epoch": 0.88073394, "global_step/max_steps": "12/13", "percentage": "92.31%", "elapsed_time": "8m 24s", "remaining_time": "42s"}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
runs/events.out.tfevents.1743269896.notebook-67532d59-da58-11ef-a92b-d22a634ff48d-0.262569.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d4bbed0eaf5ae295ce7d8ae1ac164744e3848584a1ed91e3956618e3178a5b33
|
| 3 |
+
size 10951
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 8248
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:704ebc33ac965782c742b0bc0cd5f2835ccacc8ae48ea2301abf1c4c74d4b18c
|
| 3 |
size 8248
|
val_dataset.jsonl
CHANGED
|
The diff for this file is too large to render.
See raw diff
|
|
|