Hugging Face's logo

coml
/
childlex

Safetensors
Model card Files
xet
 Community
nhamilakis commited on
Commit
42f9aec
·
verified ·
1 Parent(s): 1a02483

Upload folder using huggingface_hub

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. .gitattributes +1 -0
  2. childes/EN/adult/05/00/orthographic_gpt2/config.json +32 -0
  3. childes/EN/adult/05/00/orthographic_gpt2/generation_config.json +6 -0
  4. childes/EN/adult/05/00/orthographic_gpt2/model.safetensors +3 -0
  5. childes/EN/adult/05/00/orthographic_gpt2/training.logs +38 -0
  6. childes/EN/adult/05/00/orthographic_gpt2/training_args.bin +3 -0
  7. childes/EN/adult/05/00/orthographic_lstm/config.json +13 -0
  8. childes/EN/adult/05/00/orthographic_lstm/generation_config.json +4 -0
  9. childes/EN/adult/05/00/orthographic_lstm/model.safetensors +3 -0
  10. childes/EN/adult/05/00/orthographic_lstm/training.logs +53 -0
  11. childes/EN/adult/05/00/orthographic_lstm/training_args.bin +3 -0
  12. childes/EN/adult/15/00/orthographic_gpt2/config.json +32 -0
  13. childes/EN/adult/15/00/orthographic_gpt2/generation_config.json +6 -0
  14. childes/EN/adult/15/00/orthographic_gpt2/model.safetensors +3 -0
  15. childes/EN/adult/15/00/orthographic_gpt2/training.logs +38 -0
  16. childes/EN/adult/15/00/orthographic_gpt2/training_args.bin +3 -0
  17. childes/EN/adult/15/00/orthographic_lstm/config.json +13 -0
  18. childes/EN/adult/15/00/orthographic_lstm/generation_config.json +4 -0
  19. childes/EN/adult/15/00/orthographic_lstm/model.safetensors +3 -0
  20. childes/EN/adult/15/00/orthographic_lstm/training.logs +36 -0
  21. childes/EN/adult/15/00/orthographic_lstm/training_args.bin +3 -0
  22. childes/EN/adult/20/00/orthographic_gpt2/config.json +32 -0
  23. childes/EN/adult/20/00/orthographic_gpt2/generation_config.json +6 -0
  24. childes/EN/adult/20/00/orthographic_gpt2/model.safetensors +3 -0
  25. childes/EN/adult/20/00/orthographic_gpt2/training.logs +38 -0
  26. childes/EN/adult/20/00/orthographic_gpt2/training_args.bin +3 -0
  27. childes/EN/adult/20/00/orthographic_lstm/config.json +13 -0
  28. childes/EN/adult/20/00/orthographic_lstm/generation_config.json +4 -0
  29. childes/EN/adult/20/00/orthographic_lstm/model.safetensors +3 -0
  30. childes/EN/adult/20/00/orthographic_lstm/training.logs +36 -0
  31. childes/EN/adult/20/00/orthographic_lstm/training_args.bin +3 -0
  32. childes/EN/adult/25/00/orthographic_gpt2/config.json +32 -0
  33. childes/EN/adult/25/00/orthographic_gpt2/generation_config.json +6 -0
  34. childes/EN/adult/25/00/orthographic_gpt2/model.safetensors +3 -0
  35. childes/EN/adult/25/00/orthographic_gpt2/training.logs +38 -0
  36. childes/EN/adult/25/00/orthographic_gpt2/training_args.bin +3 -0
  37. childes/EN/adult/25/00/orthographic_lstm/config.json +13 -0
  38. childes/EN/adult/25/00/orthographic_lstm/generation_config.json +4 -0
  39. childes/EN/adult/25/00/orthographic_lstm/model.safetensors +3 -0
  40. childes/EN/adult/25/00/orthographic_lstm/training.logs +3 -0
  41. childes/EN/adult/25/00/orthographic_lstm/training_args.bin +3 -0
  42. childes/to_train.toml +29 -0
  43. stela/EN/01/00/orthographic_gpt2/config.json +32 -0
  44. stela/EN/01/00/orthographic_gpt2/generation_config.json +6 -0
  45. stela/EN/01/00/orthographic_gpt2/model.safetensors +3 -0
  46. stela/EN/01/00/orthographic_gpt2/training.logs +315 -0
  47. stela/EN/01/00/orthographic_gpt2/training_args.bin +3 -0
  48. stela/EN/01/00/orthographic_lstm/config.json +13 -0
  49. stela/EN/01/00/orthographic_lstm/model.safetensors +3 -0
  50. stela/EN/01/00/orthographic_lstm/training_args.bin +3 -0
.gitattributes CHANGED
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ childes/EN/adult/25/00/orthographic_lstm/training.logs filter=lfs diff=lfs merge=lfs -text
childes/EN/adult/05/00/orthographic_gpt2/config.json ADDED
@@ -0,0 +1,32 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation_function": "gelu_new",
3
+ "architectures": [
4
+ "GPT2LMHeadModel"
5
+ ],
6
+ "attn_pdrop": 0.1,
7
+ "bos_token_id": 50256,
8
+ "embd_pdrop": 0.1,
9
+ "eos_token_id": 50256,
10
+ "initializer_range": 0.02,
11
+ "layer_norm_epsilon": 1e-05,
12
+ "model_type": "gpt2",
13
+ "n_ctx": 1024,
14
+ "n_embd": 768,
15
+ "n_head": 12,
16
+ "n_inner": null,
17
+ "n_layer": 12,
18
+ "n_positions": 1024,
19
+ "reorder_and_upcast_attn": false,
20
+ "resid_pdrop": 0.1,
21
+ "scale_attn_by_inverse_layer_idx": false,
22
+ "scale_attn_weights": true,
23
+ "summary_activation": null,
24
+ "summary_first_dropout": 0.1,
25
+ "summary_proj_to_labels": true,
26
+ "summary_type": "cls_index",
27
+ "summary_use_proj": true,
28
+ "torch_dtype": "float32",
29
+ "transformers_version": "4.49.0",
30
+ "use_cache": true,
31
+ "vocab_size": 115
32
+ }
childes/EN/adult/05/00/orthographic_gpt2/generation_config.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 50256,
4
+ "eos_token_id": 50256,
5
+ "transformers_version": "4.49.0"
6
+ }
childes/EN/adult/05/00/orthographic_gpt2/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ccf3c666e95f031a3d4a80c072bb12095b2dfeb75d7a2abbbcedd1915f89d89b
3
+ size 343737984
childes/EN/adult/05/00/orthographic_gpt2/training.logs ADDED
@@ -0,0 +1,38 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 06/19/2025 01:08:49 - lexical_benchmark.train.run_train - INFO - Loading gpt2 trainer class.
2
+ 06/19/2025 01:09:00 - datasets - INFO - PyTorch version 2.5.1 available.
3
+ 06/19/2025 01:09:00 - datasets - INFO - Polars version 1.24.0 available.
4
+ 06/19/2025 01:09:01 - lexical_benchmark.train.train_params - INFO - Loading params from /lustre/fsn1/projects/rech/hhb/uye44va/workspace/lm_benchmark/dev/code/src/lexical_benchmark/train/train-params.toml
5
+ 06/19/2025 01:09:01 - lexical_benchmark.train.trainers.gpt2 - INFO - Loading char tokenizer
6
+ 06/19/2025 01:09:01 - lexical_benchmark.train.trainers.gpt2 - INFO - Tokenizer loaded with vocabulary size: 115
7
+ 06/19/2025 01:09:01 - lexical_benchmark.train.trainers.gpt2 - INFO - Loading and processing datasets with joined utterances (max_length=1024)
8
+ 06/19/2025 01:09:01 - lexical_benchmark.train.tokenizers - INFO - Reading text from /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/datasets/childes/by_size/EN/adult/05/00/train.tokenized
9
+ 06/19/2025 01:09:01 - lexical_benchmark.train.tokenizers - INFO - Read 561041 lines from file
10
+ 06/19/2025 01:09:21 - lexical_benchmark.train.tokenizers - INFO - Created 11852 joined examples from 561041 original lines
11
+ 06/19/2025 01:09:23 - lexical_benchmark.train.tokenizers - INFO - Reading text from /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/datasets/childes/by_size/EN/adult/dev/dev.tokenized
12
+ 06/19/2025 01:09:23 - lexical_benchmark.train.tokenizers - INFO - Read 316899 lines from file
13
+ 06/19/2025 01:09:34 - lexical_benchmark.train.tokenizers - INFO - Created 7325 joined examples from 316899 original lines
14
+ 06/19/2025 01:09:35 - lexical_benchmark.train.trainers.gpt2 - INFO - Training dataset size: 11852
15
+ 06/19/2025 01:09:35 - lexical_benchmark.train.trainers.gpt2 - INFO - Validation dataset size: 7325
16
+ 06/19/2025 01:09:35 - lexical_benchmark.train.trainers.gpt2 - INFO - Creating GPT2 configuration
17
+ 06/19/2025 01:09:36 - lexical_benchmark.train.trainers.gpt2 - INFO - Creating standard Trainer
18
+ 06/19/2025 01:09:37 - lexical_benchmark.train.run_train - INFO - Running training of childes_gpt2_EN_05_00
19
+ 06/19/2025 01:12:19 - lexical_benchmark.train.run_train - INFO - Loading gpt2 trainer class.
20
+ 06/19/2025 01:12:30 - datasets - INFO - PyTorch version 2.5.1 available.
21
+ 06/19/2025 01:12:30 - datasets - INFO - Polars version 1.24.0 available.
22
+ 06/19/2025 01:12:31 - lexical_benchmark.train.train_params - INFO - Loading params from /lustre/fsn1/projects/rech/hhb/uye44va/workspace/lm_benchmark/dev/code/src/lexical_benchmark/train/train-params.toml
23
+ 06/19/2025 01:12:31 - lexical_benchmark.train.trainers.gpt2 - INFO - Loading char tokenizer
24
+ 06/19/2025 01:12:31 - lexical_benchmark.train.trainers.gpt2 - INFO - Tokenizer loaded with vocabulary size: 115
25
+ 06/19/2025 01:12:31 - lexical_benchmark.train.trainers.gpt2 - INFO - Loading and processing datasets with joined utterances (max_length=1024)
26
+ 06/19/2025 01:12:31 - lexical_benchmark.train.tokenizers - INFO - Reading text from /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/datasets/childes/by_size/EN/adult/05/00/train.tokenized
27
+ 06/19/2025 01:12:31 - lexical_benchmark.train.tokenizers - INFO - Read 561041 lines from file
28
+ 06/19/2025 01:12:51 - lexical_benchmark.train.tokenizers - INFO - Created 11852 joined examples from 561041 original lines
29
+ 06/19/2025 01:12:53 - lexical_benchmark.train.tokenizers - INFO - Reading text from /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/datasets/childes/by_size/EN/adult/dev/dev.tokenized
30
+ 06/19/2025 01:12:53 - lexical_benchmark.train.tokenizers - INFO - Read 316899 lines from file
31
+ 06/19/2025 01:13:05 - lexical_benchmark.train.tokenizers - INFO - Created 7325 joined examples from 316899 original lines
32
+ 06/19/2025 01:13:06 - lexical_benchmark.train.trainers.gpt2 - INFO - Training dataset size: 11852
33
+ 06/19/2025 01:13:06 - lexical_benchmark.train.trainers.gpt2 - INFO - Validation dataset size: 7325
34
+ 06/19/2025 01:13:06 - lexical_benchmark.train.trainers.gpt2 - INFO - Creating GPT2 configuration
35
+ 06/19/2025 01:13:07 - lexical_benchmark.train.trainers.gpt2 - INFO - Creating standard Trainer
36
+ 06/19/2025 01:13:07 - lexical_benchmark.train.run_train - INFO - Running training of childes_gpt2_EN_05_00
37
+ 06/19/2025 02:46:25 - lexical_benchmark.train.run_train - INFO - Training of childes_gpt2_EN_05_00 has completed !
38
+ 06/19/2025 02:46:25 - lexical_benchmark.train.run_train - INFO - Model saved to /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/models/childes/EN/adult/05/00/gpt2
childes/EN/adult/05/00/orthographic_gpt2/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fab672c263a3536143e829f27276021c3a3b97db7119200922890f4e99af56c7
3
+ size 5496
childes/EN/adult/05/00/orthographic_lstm/config.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "LSTMForLanguageModeling"
4
+ ],
5
+ "dropout": 0.1,
6
+ "embedding_dim": 200,
7
+ "hidden_size": 1024,
8
+ "model_type": "lstm",
9
+ "num_layers": 3,
10
+ "torch_dtype": "float32",
11
+ "transformers_version": "4.49.0",
12
+ "vocab_size": 115
13
+ }
childes/EN/adult/05/00/orthographic_lstm/generation_config.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "transformers_version": "4.49.0"
4
+ }
childes/EN/adult/05/00/orthographic_lstm/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:986070fc467d6b2b4caef72683b4f0a9e41cdcfc5fe2d6eba7768394613778e2
3
+ size 87825996
childes/EN/adult/05/00/orthographic_lstm/training.logs ADDED
@@ -0,0 +1,53 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 06/19/2025 00:46:24 - lexical_benchmark.train.run_train - INFO - Loading lstm trainer class.
2
+ 06/19/2025 00:46:28 - datasets - INFO - PyTorch version 2.5.1 available.
3
+ 06/19/2025 00:46:28 - datasets - INFO - Polars version 1.24.0 available.
4
+ 06/19/2025 00:46:29 - lexical_benchmark.train.train_params - INFO - Loading params from /lustre/fsn1/projects/rech/hhb/uye44va/workspace/lm_benchmark/dev/code/src/lexical_benchmark/train/train-params.toml
5
+ 06/19/2025 00:46:29 - lexical_benchmark.train.trainers.lstm - INFO - Loading char-tokenizer
6
+ 06/19/2025 00:46:29 - lexical_benchmark.train.trainers.lstm - INFO - Tokenizer loaded with vocabulary size: 115
7
+ 06/19/2025 00:46:29 - lexical_benchmark.train.trainers.lstm - INFO - Loading and processing datasets with joined utterances (max_length=1024)
8
+ 06/19/2025 00:46:29 - lexical_benchmark.train.tokenizers - INFO - Reading text from /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/datasets/childes/by_size/EN/adult/05/00/train.tokenized
9
+ 06/19/2025 00:46:29 - lexical_benchmark.train.tokenizers - INFO - Read 561041 lines from file
10
+ 06/19/2025 00:46:49 - lexical_benchmark.train.tokenizers - INFO - Created 11852 joined examples from 561041 original lines
11
+ 06/19/2025 00:46:51 - lexical_benchmark.train.tokenizers - INFO - Reading text from /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/datasets/childes/by_size/EN/adult/dev/dev.tokenized
12
+ 06/19/2025 00:46:51 - lexical_benchmark.train.tokenizers - INFO - Read 316899 lines from file
13
+ 06/19/2025 00:47:02 - lexical_benchmark.train.tokenizers - INFO - Created 7325 joined examples from 316899 original lines
14
+ 06/19/2025 00:47:03 - lexical_benchmark.train.trainers.lstm - INFO - Training dataset size: 11852
15
+ 06/19/2025 00:47:03 - lexical_benchmark.train.trainers.lstm - INFO - Validation dataset size: 7325
16
+ 06/19/2025 00:47:03 - lexical_benchmark.train.trainers.lstm - INFO - Loading configurations & initialising LSTM model trainer
17
+ 06/19/2025 00:47:04 - lexical_benchmark.train.run_train - INFO - Running training of childes_lstm_EN_05_00
18
+ 06/19/2025 01:08:44 - lexical_benchmark.train.run_train - INFO - Loading lstm trainer class.
19
+ 06/19/2025 01:08:48 - datasets - INFO - PyTorch version 2.5.1 available.
20
+ 06/19/2025 01:08:48 - datasets - INFO - Polars version 1.24.0 available.
21
+ 06/19/2025 01:08:48 - lexical_benchmark.train.train_params - INFO - Loading params from /lustre/fsn1/projects/rech/hhb/uye44va/workspace/lm_benchmark/dev/code/src/lexical_benchmark/train/train-params.toml
22
+ 06/19/2025 01:08:48 - lexical_benchmark.train.trainers.lstm - INFO - Loading char-tokenizer
23
+ 06/19/2025 01:08:48 - lexical_benchmark.train.trainers.lstm - INFO - Tokenizer loaded with vocabulary size: 115
24
+ 06/19/2025 01:08:48 - lexical_benchmark.train.trainers.lstm - INFO - Loading and processing datasets with joined utterances (max_length=1024)
25
+ 06/19/2025 01:08:48 - lexical_benchmark.train.tokenizers - INFO - Reading text from /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/datasets/childes/by_size/EN/adult/05/00/train.tokenized
26
+ 06/19/2025 01:08:48 - lexical_benchmark.train.tokenizers - INFO - Read 561041 lines from file
27
+ 06/19/2025 01:09:09 - lexical_benchmark.train.tokenizers - INFO - Created 11852 joined examples from 561041 original lines
28
+ 06/19/2025 01:09:10 - lexical_benchmark.train.tokenizers - INFO - Reading text from /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/datasets/childes/by_size/EN/adult/dev/dev.tokenized
29
+ 06/19/2025 01:09:10 - lexical_benchmark.train.tokenizers - INFO - Read 316899 lines from file
30
+ 06/19/2025 01:09:22 - lexical_benchmark.train.tokenizers - INFO - Created 7325 joined examples from 316899 original lines
31
+ 06/19/2025 01:09:23 - lexical_benchmark.train.trainers.lstm - INFO - Training dataset size: 11852
32
+ 06/19/2025 01:09:23 - lexical_benchmark.train.trainers.lstm - INFO - Validation dataset size: 7325
33
+ 06/19/2025 01:09:23 - lexical_benchmark.train.trainers.lstm - INFO - Loading configurations & initialising LSTM model trainer
34
+ 06/19/2025 01:09:24 - lexical_benchmark.train.run_train - INFO - Running training of childes_lstm_EN_05_00
35
+ 06/19/2025 01:12:15 - lexical_benchmark.train.run_train - INFO - Loading lstm trainer class.
36
+ 06/19/2025 01:12:20 - datasets - INFO - PyTorch version 2.5.1 available.
37
+ 06/19/2025 01:12:20 - datasets - INFO - Polars version 1.24.0 available.
38
+ 06/19/2025 01:12:21 - lexical_benchmark.train.train_params - INFO - Loading params from /lustre/fsn1/projects/rech/hhb/uye44va/workspace/lm_benchmark/dev/code/src/lexical_benchmark/train/train-params.toml
39
+ 06/19/2025 01:12:21 - lexical_benchmark.train.trainers.lstm - INFO - Loading char-tokenizer
40
+ 06/19/2025 01:12:21 - lexical_benchmark.train.trainers.lstm - INFO - Tokenizer loaded with vocabulary size: 115
41
+ 06/19/2025 01:12:21 - lexical_benchmark.train.trainers.lstm - INFO - Loading and processing datasets with joined utterances (max_length=1024)
42
+ 06/19/2025 01:12:21 - lexical_benchmark.train.tokenizers - INFO - Reading text from /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/datasets/childes/by_size/EN/adult/05/00/train.tokenized
43
+ 06/19/2025 01:12:21 - lexical_benchmark.train.tokenizers - INFO - Read 561041 lines from file
44
+ 06/19/2025 01:12:40 - lexical_benchmark.train.tokenizers - INFO - Created 11852 joined examples from 561041 original lines
45
+ 06/19/2025 01:12:42 - lexical_benchmark.train.tokenizers - INFO - Reading text from /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/datasets/childes/by_size/EN/adult/dev/dev.tokenized
46
+ 06/19/2025 01:12:42 - lexical_benchmark.train.tokenizers - INFO - Read 316899 lines from file
47
+ 06/19/2025 01:12:54 - lexical_benchmark.train.tokenizers - INFO - Created 7325 joined examples from 316899 original lines
48
+ 06/19/2025 01:12:55 - lexical_benchmark.train.trainers.lstm - INFO - Training dataset size: 11852
49
+ 06/19/2025 01:12:55 - lexical_benchmark.train.trainers.lstm - INFO - Validation dataset size: 7325
50
+ 06/19/2025 01:12:55 - lexical_benchmark.train.trainers.lstm - INFO - Loading configurations & initialising LSTM model trainer
51
+ 06/19/2025 01:12:56 - lexical_benchmark.train.run_train - INFO - Running training of childes_lstm_EN_05_00
52
+ 06/19/2025 07:24:19 - lexical_benchmark.train.run_train - INFO - Training of childes_lstm_EN_05_00 has completed !
53
+ 06/19/2025 07:24:19 - lexical_benchmark.train.run_train - INFO - Model saved to /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/models/childes/EN/adult/05/00/lstm
childes/EN/adult/05/00/orthographic_lstm/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:988f15f68d9ca2b3ba55756b7cbc7d948011e56f06eab4f5a5758fdd9dc7241e
3
+ size 5496
childes/EN/adult/15/00/orthographic_gpt2/config.json ADDED
@@ -0,0 +1,32 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation_function": "gelu_new",
3
+ "architectures": [
4
+ "GPT2LMHeadModel"
5
+ ],
6
+ "attn_pdrop": 0.1,
7
+ "bos_token_id": 50256,
8
+ "embd_pdrop": 0.1,
9
+ "eos_token_id": 50256,
10
+ "initializer_range": 0.02,
11
+ "layer_norm_epsilon": 1e-05,
12
+ "model_type": "gpt2",
13
+ "n_ctx": 1024,
14
+ "n_embd": 768,
15
+ "n_head": 12,
16
+ "n_inner": null,
17
+ "n_layer": 12,
18
+ "n_positions": 1024,
19
+ "reorder_and_upcast_attn": false,
20
+ "resid_pdrop": 0.1,
21
+ "scale_attn_by_inverse_layer_idx": false,
22
+ "scale_attn_weights": true,
23
+ "summary_activation": null,
24
+ "summary_first_dropout": 0.1,
25
+ "summary_proj_to_labels": true,
26
+ "summary_type": "cls_index",
27
+ "summary_use_proj": true,
28
+ "torch_dtype": "float32",
29
+ "transformers_version": "4.49.0",
30
+ "use_cache": true,
31
+ "vocab_size": 115
32
+ }
childes/EN/adult/15/00/orthographic_gpt2/generation_config.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 50256,
4
+ "eos_token_id": 50256,
5
+ "transformers_version": "4.49.0"
6
+ }
childes/EN/adult/15/00/orthographic_gpt2/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c78fa75bbb5b8c628ccc71720929b7b54094bdfd2b9c7744c620437a3e85b30a
3
+ size 343737984
childes/EN/adult/15/00/orthographic_gpt2/training.logs ADDED
@@ -0,0 +1,38 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 06/19/2025 01:08:49 - lexical_benchmark.train.run_train - INFO - Loading gpt2 trainer class.
2
+ 06/19/2025 01:09:00 - datasets - INFO - PyTorch version 2.5.1 available.
3
+ 06/19/2025 01:09:00 - datasets - INFO - Polars version 1.24.0 available.
4
+ 06/19/2025 01:09:01 - lexical_benchmark.train.train_params - INFO - Loading params from /lustre/fsn1/projects/rech/hhb/uye44va/workspace/lm_benchmark/dev/code/src/lexical_benchmark/train/train-params.toml
5
+ 06/19/2025 01:09:01 - lexical_benchmark.train.trainers.gpt2 - INFO - Loading char tokenizer
6
+ 06/19/2025 01:09:01 - lexical_benchmark.train.trainers.gpt2 - INFO - Tokenizer loaded with vocabulary size: 115
7
+ 06/19/2025 01:09:01 - lexical_benchmark.train.trainers.gpt2 - INFO - Loading and processing datasets with joined utterances (max_length=1024)
8
+ 06/19/2025 01:09:03 - lexical_benchmark.train.tokenizers - INFO - Reading text from /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/datasets/childes/by_size/EN/adult/15/00/train.tokenized
9
+ 06/19/2025 01:09:03 - lexical_benchmark.train.tokenizers - INFO - Read 1633721 lines from file
10
+ 06/19/2025 01:10:01 - lexical_benchmark.train.tokenizers - INFO - Created 35893 joined examples from 1633721 original lines
11
+ 06/19/2025 01:10:07 - lexical_benchmark.train.tokenizers - INFO - Reading text from /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/datasets/childes/by_size/EN/adult/dev/dev.tokenized
12
+ 06/19/2025 01:10:07 - lexical_benchmark.train.tokenizers - INFO - Read 316899 lines from file
13
+ 06/19/2025 01:10:19 - lexical_benchmark.train.tokenizers - INFO - Created 7325 joined examples from 316899 original lines
14
+ 06/19/2025 01:10:20 - lexical_benchmark.train.trainers.gpt2 - INFO - Training dataset size: 35893
15
+ 06/19/2025 01:10:20 - lexical_benchmark.train.trainers.gpt2 - INFO - Validation dataset size: 7325
16
+ 06/19/2025 01:10:20 - lexical_benchmark.train.trainers.gpt2 - INFO - Creating GPT2 configuration
17
+ 06/19/2025 01:10:20 - lexical_benchmark.train.trainers.gpt2 - INFO - Creating standard Trainer
18
+ 06/19/2025 01:10:21 - lexical_benchmark.train.run_train - INFO - Running training of childes_gpt2_EN_15_00
19
+ 06/19/2025 01:12:19 - lexical_benchmark.train.run_train - INFO - Loading gpt2 trainer class.
20
+ 06/19/2025 01:12:30 - datasets - INFO - PyTorch version 2.5.1 available.
21
+ 06/19/2025 01:12:30 - datasets - INFO - Polars version 1.24.0 available.
22
+ 06/19/2025 01:12:31 - lexical_benchmark.train.train_params - INFO - Loading params from /lustre/fsn1/projects/rech/hhb/uye44va/workspace/lm_benchmark/dev/code/src/lexical_benchmark/train/train-params.toml
23
+ 06/19/2025 01:12:31 - lexical_benchmark.train.trainers.gpt2 - INFO - Loading char tokenizer
24
+ 06/19/2025 01:12:31 - lexical_benchmark.train.trainers.gpt2 - INFO - Tokenizer loaded with vocabulary size: 115
25
+ 06/19/2025 01:12:31 - lexical_benchmark.train.trainers.gpt2 - INFO - Loading and processing datasets with joined utterances (max_length=1024)
26
+ 06/19/2025 01:12:31 - lexical_benchmark.train.tokenizers - INFO - Reading text from /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/datasets/childes/by_size/EN/adult/15/00/train.tokenized
27
+ 06/19/2025 01:12:31 - lexical_benchmark.train.tokenizers - INFO - Read 1633721 lines from file
28
+ 06/19/2025 01:13:30 - lexical_benchmark.train.tokenizers - INFO - Created 35893 joined examples from 1633721 original lines
29
+ 06/19/2025 01:13:36 - lexical_benchmark.train.tokenizers - INFO - Reading text from /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/datasets/childes/by_size/EN/adult/dev/dev.tokenized
30
+ 06/19/2025 01:13:36 - lexical_benchmark.train.tokenizers - INFO - Read 316899 lines from file
31
+ 06/19/2025 01:13:47 - lexical_benchmark.train.tokenizers - INFO - Created 7325 joined examples from 316899 original lines
32
+ 06/19/2025 01:13:49 - lexical_benchmark.train.trainers.gpt2 - INFO - Training dataset size: 35893
33
+ 06/19/2025 01:13:49 - lexical_benchmark.train.trainers.gpt2 - INFO - Validation dataset size: 7325
34
+ 06/19/2025 01:13:49 - lexical_benchmark.train.trainers.gpt2 - INFO - Creating GPT2 configuration
35
+ 06/19/2025 01:13:49 - lexical_benchmark.train.trainers.gpt2 - INFO - Creating standard Trainer
36
+ 06/19/2025 01:13:51 - lexical_benchmark.train.run_train - INFO - Running training of childes_gpt2_EN_15_00
37
+ 06/19/2025 03:49:45 - lexical_benchmark.train.run_train - INFO - Training of childes_gpt2_EN_15_00 has completed !
38
+ 06/19/2025 03:49:46 - lexical_benchmark.train.run_train - INFO - Model saved to /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/models/childes/EN/adult/15/00/gpt2
childes/EN/adult/15/00/orthographic_gpt2/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b001e0beba0c81b208a01ee4d6b4f4765d25b360565fe569ed7c1ae870ece854
3
+ size 5496
childes/EN/adult/15/00/orthographic_lstm/config.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "LSTMForLanguageModeling"
4
+ ],
5
+ "dropout": 0.1,
6
+ "embedding_dim": 200,
7
+ "hidden_size": 1024,
8
+ "model_type": "lstm",
9
+ "num_layers": 3,
10
+ "torch_dtype": "float32",
11
+ "transformers_version": "4.49.0",
12
+ "vocab_size": 115
13
+ }
childes/EN/adult/15/00/orthographic_lstm/generation_config.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "transformers_version": "4.49.0"
4
+ }
childes/EN/adult/15/00/orthographic_lstm/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fd91a88444c8a4050355e4928d7c1d1d808698d1a3e252ec99c4908276239362
3
+ size 87825996
childes/EN/adult/15/00/orthographic_lstm/training.logs ADDED
@@ -0,0 +1,36 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 06/19/2025 01:08:49 - lexical_benchmark.train.run_train - INFO - Loading lstm trainer class.
2
+ 06/19/2025 01:09:00 - datasets - INFO - PyTorch version 2.5.1 available.
3
+ 06/19/2025 01:09:00 - datasets - INFO - Polars version 1.24.0 available.
4
+ 06/19/2025 01:09:01 - lexical_benchmark.train.train_params - INFO - Loading params from /lustre/fsn1/projects/rech/hhb/uye44va/workspace/lm_benchmark/dev/code/src/lexical_benchmark/train/train-params.toml
5
+ 06/19/2025 01:09:01 - lexical_benchmark.train.trainers.lstm - INFO - Loading char-tokenizer
6
+ 06/19/2025 01:09:01 - lexical_benchmark.train.trainers.lstm - INFO - Tokenizer loaded with vocabulary size: 115
7
+ 06/19/2025 01:09:01 - lexical_benchmark.train.trainers.lstm - INFO - Loading and processing datasets with joined utterances (max_length=1024)
8
+ 06/19/2025 01:09:03 - lexical_benchmark.train.tokenizers - INFO - Reading text from /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/datasets/childes/by_size/EN/adult/15/00/train.tokenized
9
+ 06/19/2025 01:09:03 - lexical_benchmark.train.tokenizers - INFO - Read 1633721 lines from file
10
+ 06/19/2025 01:10:01 - lexical_benchmark.train.tokenizers - INFO - Created 35893 joined examples from 1633721 original lines
11
+ 06/19/2025 01:10:07 - lexical_benchmark.train.tokenizers - INFO - Reading text from /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/datasets/childes/by_size/EN/adult/dev/dev.tokenized
12
+ 06/19/2025 01:10:07 - lexical_benchmark.train.tokenizers - INFO - Read 316899 lines from file
13
+ 06/19/2025 01:10:19 - lexical_benchmark.train.tokenizers - INFO - Created 7325 joined examples from 316899 original lines
14
+ 06/19/2025 01:10:20 - lexical_benchmark.train.trainers.lstm - INFO - Training dataset size: 35893
15
+ 06/19/2025 01:10:20 - lexical_benchmark.train.trainers.lstm - INFO - Validation dataset size: 7325
16
+ 06/19/2025 01:10:20 - lexical_benchmark.train.trainers.lstm - INFO - Loading configurations & initialising LSTM model trainer
17
+ 06/19/2025 01:10:20 - lexical_benchmark.train.run_train - INFO - Running training of childes_lstm_EN_15_00
18
+ 06/19/2025 01:12:19 - lexical_benchmark.train.run_train - INFO - Loading lstm trainer class.
19
+ 06/19/2025 01:12:30 - datasets - INFO - PyTorch version 2.5.1 available.
20
+ 06/19/2025 01:12:30 - datasets - INFO - Polars version 1.24.0 available.
21
+ 06/19/2025 01:12:31 - lexical_benchmark.train.train_params - INFO - Loading params from /lustre/fsn1/projects/rech/hhb/uye44va/workspace/lm_benchmark/dev/code/src/lexical_benchmark/train/train-params.toml
22
+ 06/19/2025 01:12:31 - lexical_benchmark.train.trainers.lstm - INFO - Loading char-tokenizer
23
+ 06/19/2025 01:12:31 - lexical_benchmark.train.trainers.lstm - INFO - Tokenizer loaded with vocabulary size: 115
24
+ 06/19/2025 01:12:31 - lexical_benchmark.train.trainers.lstm - INFO - Loading and processing datasets with joined utterances (max_length=1024)
25
+ 06/19/2025 01:12:31 - lexical_benchmark.train.tokenizers - INFO - Reading text from /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/datasets/childes/by_size/EN/adult/15/00/train.tokenized
26
+ 06/19/2025 01:12:31 - lexical_benchmark.train.tokenizers - INFO - Read 1633721 lines from file
27
+ 06/19/2025 01:13:30 - lexical_benchmark.train.tokenizers - INFO - Created 35893 joined examples from 1633721 original lines
28
+ 06/19/2025 01:13:36 - lexical_benchmark.train.tokenizers - INFO - Reading text from /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/datasets/childes/by_size/EN/adult/dev/dev.tokenized
29
+ 06/19/2025 01:13:36 - lexical_benchmark.train.tokenizers - INFO - Read 316899 lines from file
30
+ 06/19/2025 01:13:48 - lexical_benchmark.train.tokenizers - INFO - Created 7325 joined examples from 316899 original lines
31
+ 06/19/2025 01:13:49 - lexical_benchmark.train.trainers.lstm - INFO - Training dataset size: 35893
32
+ 06/19/2025 01:13:49 - lexical_benchmark.train.trainers.lstm - INFO - Validation dataset size: 7325
33
+ 06/19/2025 01:13:49 - lexical_benchmark.train.trainers.lstm - INFO - Loading configurations & initialising LSTM model trainer
34
+ 06/19/2025 01:13:51 - lexical_benchmark.train.run_train - INFO - Running training of childes_lstm_EN_15_00
35
+ 06/19/2025 12:29:35 - lexical_benchmark.train.run_train - INFO - Training of childes_lstm_EN_15_00 has completed !
36
+ 06/19/2025 12:29:35 - lexical_benchmark.train.run_train - INFO - Model saved to /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/models/childes/EN/adult/15/00/lstm
childes/EN/adult/15/00/orthographic_lstm/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:317eef37ac361a92a1b321eb000eaafa336c3e81c0898947746270ae8cc5905c
3
+ size 5496
childes/EN/adult/20/00/orthographic_gpt2/config.json ADDED
@@ -0,0 +1,32 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation_function": "gelu_new",
3
+ "architectures": [
4
+ "GPT2LMHeadModel"
5
+ ],
6
+ "attn_pdrop": 0.1,
7
+ "bos_token_id": 50256,
8
+ "embd_pdrop": 0.1,
9
+ "eos_token_id": 50256,
10
+ "initializer_range": 0.02,
11
+ "layer_norm_epsilon": 1e-05,
12
+ "model_type": "gpt2",
13
+ "n_ctx": 1024,
14
+ "n_embd": 768,
15
+ "n_head": 12,
16
+ "n_inner": null,
17
+ "n_layer": 12,
18
+ "n_positions": 1024,
19
+ "reorder_and_upcast_attn": false,
20
+ "resid_pdrop": 0.1,
21
+ "scale_attn_by_inverse_layer_idx": false,
22
+ "scale_attn_weights": true,
23
+ "summary_activation": null,
24
+ "summary_first_dropout": 0.1,
25
+ "summary_proj_to_labels": true,
26
+ "summary_type": "cls_index",
27
+ "summary_use_proj": true,
28
+ "torch_dtype": "float32",
29
+ "transformers_version": "4.49.0",
30
+ "use_cache": true,
31
+ "vocab_size": 115
32
+ }
childes/EN/adult/20/00/orthographic_gpt2/generation_config.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 50256,
4
+ "eos_token_id": 50256,
5
+ "transformers_version": "4.49.0"
6
+ }
childes/EN/adult/20/00/orthographic_gpt2/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e3e50cf3b71e8b75e0fa39345f424aa9fcbcbe85e9af12861bc31d36d3faae05
3
+ size 343737984
childes/EN/adult/20/00/orthographic_gpt2/training.logs ADDED
@@ -0,0 +1,38 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 06/19/2025 01:08:48 - lexical_benchmark.train.run_train - INFO - Loading gpt2 trainer class.
2
+ 06/19/2025 01:08:59 - datasets - INFO - PyTorch version 2.5.1 available.
3
+ 06/19/2025 01:08:59 - datasets - INFO - Polars version 1.24.0 available.
4
+ 06/19/2025 01:09:00 - lexical_benchmark.train.train_params - INFO - Loading params from /lustre/fsn1/projects/rech/hhb/uye44va/workspace/lm_benchmark/dev/code/src/lexical_benchmark/train/train-params.toml
5
+ 06/19/2025 01:09:00 - lexical_benchmark.train.trainers.gpt2 - INFO - Loading char tokenizer
6
+ 06/19/2025 01:09:00 - lexical_benchmark.train.trainers.gpt2 - INFO - Tokenizer loaded with vocabulary size: 115
7
+ 06/19/2025 01:09:00 - lexical_benchmark.train.trainers.gpt2 - INFO - Loading and processing datasets with joined utterances (max_length=1024)
8
+ 06/19/2025 01:09:02 - lexical_benchmark.train.tokenizers - INFO - Reading text from /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/datasets/childes/by_size/EN/adult/20/00/train.tokenized
9
+ 06/19/2025 01:09:03 - lexical_benchmark.train.tokenizers - INFO - Read 2254077 lines from file
10
+ 06/19/2025 01:10:24 - lexical_benchmark.train.tokenizers - INFO - Created 47509 joined examples from 2254077 original lines
11
+ 06/19/2025 01:10:31 - lexical_benchmark.train.tokenizers - INFO - Reading text from /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/datasets/childes/by_size/EN/adult/dev/dev.tokenized
12
+ 06/19/2025 01:10:31 - lexical_benchmark.train.tokenizers - INFO - Read 316899 lines from file
13
+ 06/19/2025 01:10:43 - lexical_benchmark.train.tokenizers - INFO - Created 7325 joined examples from 316899 original lines
14
+ 06/19/2025 01:10:44 - lexical_benchmark.train.trainers.gpt2 - INFO - Training dataset size: 47509
15
+ 06/19/2025 01:10:44 - lexical_benchmark.train.trainers.gpt2 - INFO - Validation dataset size: 7325
16
+ 06/19/2025 01:10:44 - lexical_benchmark.train.trainers.gpt2 - INFO - Creating GPT2 configuration
17
+ 06/19/2025 01:10:45 - lexical_benchmark.train.trainers.gpt2 - INFO - Creating standard Trainer
18
+ 06/19/2025 01:10:46 - lexical_benchmark.train.run_train - INFO - Running training of childes_gpt2_EN_20_00
19
+ 06/19/2025 01:12:19 - lexical_benchmark.train.run_train - INFO - Loading gpt2 trainer class.
20
+ 06/19/2025 01:12:29 - datasets - INFO - PyTorch version 2.5.1 available.
21
+ 06/19/2025 01:12:29 - datasets - INFO - Polars version 1.24.0 available.
22
+ 06/19/2025 01:12:30 - lexical_benchmark.train.train_params - INFO - Loading params from /lustre/fsn1/projects/rech/hhb/uye44va/workspace/lm_benchmark/dev/code/src/lexical_benchmark/train/train-params.toml
23
+ 06/19/2025 01:12:30 - lexical_benchmark.train.trainers.gpt2 - INFO - Loading char tokenizer
24
+ 06/19/2025 01:12:30 - lexical_benchmark.train.trainers.gpt2 - INFO - Tokenizer loaded with vocabulary size: 115
25
+ 06/19/2025 01:12:30 - lexical_benchmark.train.trainers.gpt2 - INFO - Loading and processing datasets with joined utterances (max_length=1024)
26
+ 06/19/2025 01:12:30 - lexical_benchmark.train.tokenizers - INFO - Reading text from /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/datasets/childes/by_size/EN/adult/20/00/train.tokenized
27
+ 06/19/2025 01:12:30 - lexical_benchmark.train.tokenizers - INFO - Read 2254077 lines from file
28
+ 06/19/2025 01:13:53 - lexical_benchmark.train.tokenizers - INFO - Created 47509 joined examples from 2254077 original lines
29
+ 06/19/2025 01:14:01 - lexical_benchmark.train.tokenizers - INFO - Reading text from /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/datasets/childes/by_size/EN/adult/dev/dev.tokenized
30
+ 06/19/2025 01:14:01 - lexical_benchmark.train.tokenizers - INFO - Read 316899 lines from file
31
+ 06/19/2025 01:14:13 - lexical_benchmark.train.tokenizers - INFO - Created 7325 joined examples from 316899 original lines
32
+ 06/19/2025 01:14:14 - lexical_benchmark.train.trainers.gpt2 - INFO - Training dataset size: 47509
33
+ 06/19/2025 01:14:14 - lexical_benchmark.train.trainers.gpt2 - INFO - Validation dataset size: 7325
34
+ 06/19/2025 01:14:14 - lexical_benchmark.train.trainers.gpt2 - INFO - Creating GPT2 configuration
35
+ 06/19/2025 01:14:15 - lexical_benchmark.train.trainers.gpt2 - INFO - Creating standard Trainer
36
+ 06/19/2025 01:14:16 - lexical_benchmark.train.run_train - INFO - Running training of childes_gpt2_EN_20_00
37
+ 06/19/2025 04:20:49 - lexical_benchmark.train.run_train - INFO - Training of childes_gpt2_EN_20_00 has completed !
38
+ 06/19/2025 04:20:49 - lexical_benchmark.train.run_train - INFO - Model saved to /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/models/childes/EN/adult/20/00/gpt2
childes/EN/adult/20/00/orthographic_gpt2/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f747f444d9b2de50670f2fe4a1fa44e86de494e73b3110a36cbba1e060316592
3
+ size 5496
childes/EN/adult/20/00/orthographic_lstm/config.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "LSTMForLanguageModeling"
4
+ ],
5
+ "dropout": 0.1,
6
+ "embedding_dim": 200,
7
+ "hidden_size": 1024,
8
+ "model_type": "lstm",
9
+ "num_layers": 3,
10
+ "torch_dtype": "float32",
11
+ "transformers_version": "4.49.0",
12
+ "vocab_size": 115
13
+ }
childes/EN/adult/20/00/orthographic_lstm/generation_config.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "transformers_version": "4.49.0"
4
+ }
childes/EN/adult/20/00/orthographic_lstm/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ac0bf5262860476e6789d19db0f6fc9547fd778d2cc3ba0010a8fadf6118b9fc
3
+ size 87825996
childes/EN/adult/20/00/orthographic_lstm/training.logs ADDED
@@ -0,0 +1,36 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 06/19/2025 01:08:49 - lexical_benchmark.train.run_train - INFO - Loading lstm trainer class.
2
+ 06/19/2025 01:09:00 - datasets - INFO - PyTorch version 2.5.1 available.
3
+ 06/19/2025 01:09:00 - datasets - INFO - Polars version 1.24.0 available.
4
+ 06/19/2025 01:09:01 - lexical_benchmark.train.train_params - INFO - Loading params from /lustre/fsn1/projects/rech/hhb/uye44va/workspace/lm_benchmark/dev/code/src/lexical_benchmark/train/train-params.toml
5
+ 06/19/2025 01:09:01 - lexical_benchmark.train.trainers.lstm - INFO - Loading char-tokenizer
6
+ 06/19/2025 01:09:01 - lexical_benchmark.train.trainers.lstm - INFO - Tokenizer loaded with vocabulary size: 115
7
+ 06/19/2025 01:09:01 - lexical_benchmark.train.trainers.lstm - INFO - Loading and processing datasets with joined utterances (max_length=1024)
8
+ 06/19/2025 01:09:03 - lexical_benchmark.train.tokenizers - INFO - Reading text from /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/datasets/childes/by_size/EN/adult/20/00/train.tokenized
9
+ 06/19/2025 01:09:04 - lexical_benchmark.train.tokenizers - INFO - Read 2254077 lines from file
10
+ 06/19/2025 01:10:23 - lexical_benchmark.train.tokenizers - INFO - Created 47509 joined examples from 2254077 original lines
11
+ 06/19/2025 01:10:31 - lexical_benchmark.train.tokenizers - INFO - Reading text from /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/datasets/childes/by_size/EN/adult/dev/dev.tokenized
12
+ 06/19/2025 01:10:31 - lexical_benchmark.train.tokenizers - INFO - Read 316899 lines from file
13
+ 06/19/2025 01:10:42 - lexical_benchmark.train.tokenizers - INFO - Created 7325 joined examples from 316899 original lines
14
+ 06/19/2025 01:10:44 - lexical_benchmark.train.trainers.lstm - INFO - Training dataset size: 47509
15
+ 06/19/2025 01:10:44 - lexical_benchmark.train.trainers.lstm - INFO - Validation dataset size: 7325
16
+ 06/19/2025 01:10:44 - lexical_benchmark.train.trainers.lstm - INFO - Loading configurations & initialising LSTM model trainer
17
+ 06/19/2025 01:10:44 - lexical_benchmark.train.run_train - INFO - Running training of childes_lstm_EN_20_00
18
+ 06/19/2025 01:12:19 - lexical_benchmark.train.run_train - INFO - Loading lstm trainer class.
19
+ 06/19/2025 01:12:29 - datasets - INFO - PyTorch version 2.5.1 available.
20
+ 06/19/2025 01:12:29 - datasets - INFO - Polars version 1.24.0 available.
21
+ 06/19/2025 01:12:30 - lexical_benchmark.train.train_params - INFO - Loading params from /lustre/fsn1/projects/rech/hhb/uye44va/workspace/lm_benchmark/dev/code/src/lexical_benchmark/train/train-params.toml
22
+ 06/19/2025 01:12:30 - lexical_benchmark.train.trainers.lstm - INFO - Loading char-tokenizer
23
+ 06/19/2025 01:12:30 - lexical_benchmark.train.trainers.lstm - INFO - Tokenizer loaded with vocabulary size: 115
24
+ 06/19/2025 01:12:30 - lexical_benchmark.train.trainers.lstm - INFO - Loading and processing datasets with joined utterances (max_length=1024)
25
+ 06/19/2025 01:12:30 - lexical_benchmark.train.tokenizers - INFO - Reading text from /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/datasets/childes/by_size/EN/adult/20/00/train.tokenized
26
+ 06/19/2025 01:12:30 - lexical_benchmark.train.tokenizers - INFO - Read 2254077 lines from file
27
+ 06/19/2025 01:13:52 - lexical_benchmark.train.tokenizers - INFO - Created 47509 joined examples from 2254077 original lines
28
+ 06/19/2025 01:14:00 - lexical_benchmark.train.tokenizers - INFO - Reading text from /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/datasets/childes/by_size/EN/adult/dev/dev.tokenized
29
+ 06/19/2025 01:14:00 - lexical_benchmark.train.tokenizers - INFO - Read 316899 lines from file
30
+ 06/19/2025 01:14:12 - lexical_benchmark.train.tokenizers - INFO - Created 7325 joined examples from 316899 original lines
31
+ 06/19/2025 01:14:13 - lexical_benchmark.train.trainers.lstm - INFO - Training dataset size: 47509
32
+ 06/19/2025 01:14:13 - lexical_benchmark.train.trainers.lstm - INFO - Validation dataset size: 7325
33
+ 06/19/2025 01:14:13 - lexical_benchmark.train.trainers.lstm - INFO - Loading configurations & initialising LSTM model trainer
34
+ 06/19/2025 01:14:14 - lexical_benchmark.train.run_train - INFO - Running training of childes_lstm_EN_20_00
35
+ 06/19/2025 12:43:16 - lexical_benchmark.train.run_train - INFO - Training of childes_lstm_EN_20_00 has completed !
36
+ 06/19/2025 12:43:16 - lexical_benchmark.train.run_train - INFO - Model saved to /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/models/childes/EN/adult/20/00/lstm
childes/EN/adult/20/00/orthographic_lstm/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dfd3dc6bcf775b415729eb17cd094c0d97edd266cfdd9b8c72342fce910bfe1d
3
+ size 5496
childes/EN/adult/25/00/orthographic_gpt2/config.json ADDED
@@ -0,0 +1,32 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation_function": "gelu_new",
3
+ "architectures": [
4
+ "GPT2LMHeadModel"
5
+ ],
6
+ "attn_pdrop": 0.1,
7
+ "bos_token_id": 50256,
8
+ "embd_pdrop": 0.1,
9
+ "eos_token_id": 50256,
10
+ "initializer_range": 0.02,
11
+ "layer_norm_epsilon": 1e-05,
12
+ "model_type": "gpt2",
13
+ "n_ctx": 1024,
14
+ "n_embd": 768,
15
+ "n_head": 12,
16
+ "n_inner": null,
17
+ "n_layer": 12,
18
+ "n_positions": 1024,
19
+ "reorder_and_upcast_attn": false,
20
+ "resid_pdrop": 0.1,
21
+ "scale_attn_by_inverse_layer_idx": false,
22
+ "scale_attn_weights": true,
23
+ "summary_activation": null,
24
+ "summary_first_dropout": 0.1,
25
+ "summary_proj_to_labels": true,
26
+ "summary_type": "cls_index",
27
+ "summary_use_proj": true,
28
+ "torch_dtype": "float32",
29
+ "transformers_version": "4.49.0",
30
+ "use_cache": true,
31
+ "vocab_size": 115
32
+ }
childes/EN/adult/25/00/orthographic_gpt2/generation_config.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 50256,
4
+ "eos_token_id": 50256,
5
+ "transformers_version": "4.49.0"
6
+ }
childes/EN/adult/25/00/orthographic_gpt2/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d23de09ee8d83dc39a0ee9f3d3d36e7ad26fc3da057613aca7ad6e319864463b
3
+ size 343737984
childes/EN/adult/25/00/orthographic_gpt2/training.logs ADDED
@@ -0,0 +1,38 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 06/16/2025 19:08:05 - lexical_benchmark.train.run_train - INFO - Loading gpt2 trainer class.
2
+ 06/16/2025 19:08:09 - datasets - INFO - PyTorch version 2.5.1 available.
3
+ 06/16/2025 19:08:09 - datasets - INFO - Polars version 1.24.0 available.
4
+ 06/16/2025 19:08:09 - lexical_benchmark.train.train_params - INFO - Loading params from /lustre/fsn1/projects/rech/hhb/uye44va/workspace/lm_benchmark/dev/code/src/lexical_benchmark/train/train-params.toml
5
+ 06/16/2025 19:08:09 - lexical_benchmark.train.trainers.gpt2 - INFO - Loading char tokenizer
6
+ 06/16/2025 19:08:09 - lexical_benchmark.train.trainers.gpt2 - INFO - Tokenizer loaded with vocabulary size: 115
7
+ 06/16/2025 19:08:09 - lexical_benchmark.train.trainers.gpt2 - INFO - Loading and processing datasets with joined utterances (max_length=1024)
8
+ 06/16/2025 19:08:09 - lexical_benchmark.train.tokenizers - INFO - Reading text from /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/datasets/childes/txt/adult/EN.tokenized.train.txt
9
+ 06/16/2025 19:08:10 - lexical_benchmark.train.tokenizers - INFO - Read 3381273 lines from file
10
+ 06/16/2025 19:10:19 - lexical_benchmark.train.tokenizers - INFO - Created 71947 joined examples from 3381273 original lines
11
+ 06/16/2025 19:10:31 - lexical_benchmark.train.tokenizers - INFO - Reading text from /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/datasets/childes/txt/adult/EN.tokenized.dev.txt
12
+ 06/16/2025 19:10:31 - lexical_benchmark.train.tokenizers - INFO - Read 353438 lines from file
13
+ 06/16/2025 19:10:44 - lexical_benchmark.train.tokenizers - INFO - Created 6552 joined examples from 353438 original lines
14
+ 06/16/2025 19:10:45 - lexical_benchmark.train.trainers.gpt2 - INFO - Training dataset size: 71947
15
+ 06/16/2025 19:10:45 - lexical_benchmark.train.trainers.gpt2 - INFO - Validation dataset size: 6552
16
+ 06/16/2025 19:10:45 - lexical_benchmark.train.trainers.gpt2 - INFO - Creating GPT2 configuration
17
+ 06/16/2025 19:10:46 - lexical_benchmark.train.trainers.gpt2 - INFO - Creating standard Trainer
18
+ 06/16/2025 19:10:47 - lexical_benchmark.train.run_train - INFO - Running training of childes_gpt2_childes_EN_adult
19
+ 06/16/2025 19:24:45 - lexical_benchmark.train.run_train - INFO - Loading gpt2 trainer class.
20
+ 06/16/2025 19:24:55 - datasets - INFO - PyTorch version 2.5.1 available.
21
+ 06/16/2025 19:24:55 - datasets - INFO - Polars version 1.24.0 available.
22
+ 06/16/2025 19:24:56 - lexical_benchmark.train.train_params - INFO - Loading params from /lustre/fsn1/projects/rech/hhb/uye44va/workspace/lm_benchmark/dev/code/src/lexical_benchmark/train/train-params.toml
23
+ 06/16/2025 19:24:56 - lexical_benchmark.train.trainers.gpt2 - INFO - Loading char tokenizer
24
+ 06/16/2025 19:24:56 - lexical_benchmark.train.trainers.gpt2 - INFO - Tokenizer loaded with vocabulary size: 115
25
+ 06/16/2025 19:24:56 - lexical_benchmark.train.trainers.gpt2 - INFO - Loading and processing datasets with joined utterances (max_length=1024)
26
+ 06/16/2025 19:24:56 - lexical_benchmark.train.tokenizers - INFO - Reading text from /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/datasets/childes/txt/adult/EN.tokenized.train.txt
27
+ 06/16/2025 19:24:56 - lexical_benchmark.train.tokenizers - INFO - Read 3381273 lines from file
28
+ 06/16/2025 19:26:57 - lexical_benchmark.train.tokenizers - INFO - Created 71947 joined examples from 3381273 original lines
29
+ 06/16/2025 19:27:09 - lexical_benchmark.train.tokenizers - INFO - Reading text from /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/datasets/childes/txt/adult/EN.tokenized.dev.txt
30
+ 06/16/2025 19:27:09 - lexical_benchmark.train.tokenizers - INFO - Read 353438 lines from file
31
+ 06/16/2025 19:27:20 - lexical_benchmark.train.tokenizers - INFO - Created 6552 joined examples from 353438 original lines
32
+ 06/16/2025 19:27:21 - lexical_benchmark.train.trainers.gpt2 - INFO - Training dataset size: 71947
33
+ 06/16/2025 19:27:21 - lexical_benchmark.train.trainers.gpt2 - INFO - Validation dataset size: 6552
34
+ 06/16/2025 19:27:21 - lexical_benchmark.train.trainers.gpt2 - INFO - Creating GPT2 configuration
35
+ 06/16/2025 19:27:22 - lexical_benchmark.train.trainers.gpt2 - INFO - Creating standard Trainer
36
+ 06/16/2025 19:27:25 - lexical_benchmark.train.run_train - INFO - Running training of childes_gpt2_childes_EN_adult
37
+ 06/17/2025 04:05:49 - lexical_benchmark.train.run_train - INFO - Training of childes_gpt2_childes_EN_adult has completed !
38
+ 06/17/2025 04:05:50 - lexical_benchmark.train.run_train - INFO - Model saved to /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/models/childes/adult/gpt2
childes/EN/adult/25/00/orthographic_gpt2/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:43d538a1608c1c6aed0a01654ac05e1a8424269b12596a7eb79f5a9d41bd8759
3
+ size 5496
childes/EN/adult/25/00/orthographic_lstm/config.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "LSTMForLanguageModeling"
4
+ ],
5
+ "dropout": 0.1,
6
+ "embedding_dim": 200,
7
+ "hidden_size": 1024,
8
+ "model_type": "lstm",
9
+ "num_layers": 3,
10
+ "torch_dtype": "float32",
11
+ "transformers_version": "4.49.0",
12
+ "vocab_size": 115
13
+ }
childes/EN/adult/25/00/orthographic_lstm/generation_config.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "transformers_version": "4.49.0"
4
+ }
childes/EN/adult/25/00/orthographic_lstm/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3dce5ea2d975b1e64f78caae9f8223780c45dce083af4d339a88986cf270724c
3
+ size 87825996
childes/EN/adult/25/00/orthographic_lstm/training.logs ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6e44294d22cfd922d67a7c1a2ccd9742c3d829f949b8aac6000d668a2d612073
3
+ size 85801225
childes/EN/adult/25/00/orthographic_lstm/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6b7b8fcef6a2739d8b335014bff58c5a40c48adb2f6e99e549420794f7f3bccf
3
+ size 5496
childes/to_train.toml ADDED
@@ -0,0 +1,29 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [index.0]
2
+ model_type = 'lstm'
3
+ lang = 'EN'
4
+ split = '10'
5
+ chunk = '00'
6
+ dataset_name = 'childes'
7
+ completed_training = false
8
+ generation_root = '/lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/generations/checkpoints.clean/childes/adult/EN/01/00/lstm'
9
+ model_root = '/lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/models/childes/adult/EN/01/00/lstm'
10
+ last_checkpoint = 'null'
11
+ resume = true
12
+ override = false
13
+ resume_id = 'null'
14
+ batch_size = 128
15
+
16
+ [index.1]
17
+ model_type = 'gpt2'
18
+ lang = 'EN'
19
+ split = '10'
20
+ chunk = '00'
21
+ dataset_name = 'childes'
22
+ completed_training = false
23
+ generation_root = '/lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/generations/checkpoints.clean/childes/adult/EN/01/00/gpt2'
24
+ model_root = '/lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/models/childes/adult/EN/01/00/gpt2'
25
+ last_checkpoint = 'null'
26
+ resume = true
27
+ override = false
28
+ resume_id = 'null'
29
+ batch_size = 32
stela/EN/01/00/orthographic_gpt2/config.json ADDED
@@ -0,0 +1,32 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation_function": "gelu_new",
3
+ "architectures": [
4
+ "GPT2LMHeadModel"
5
+ ],
6
+ "attn_pdrop": 0.1,
7
+ "bos_token_id": 50256,
8
+ "embd_pdrop": 0.1,
9
+ "eos_token_id": 50256,
10
+ "initializer_range": 0.02,
11
+ "layer_norm_epsilon": 1e-05,
12
+ "model_type": "gpt2",
13
+ "n_ctx": 1024,
14
+ "n_embd": 768,
15
+ "n_head": 12,
16
+ "n_inner": null,
17
+ "n_layer": 12,
18
+ "n_positions": 1024,
19
+ "reorder_and_upcast_attn": false,
20
+ "resid_pdrop": 0.1,
21
+ "scale_attn_by_inverse_layer_idx": false,
22
+ "scale_attn_weights": true,
23
+ "summary_activation": null,
24
+ "summary_first_dropout": 0.1,
25
+ "summary_proj_to_labels": true,
26
+ "summary_type": "cls_index",
27
+ "summary_use_proj": true,
28
+ "torch_dtype": "float32",
29
+ "transformers_version": "4.49.0",
30
+ "use_cache": true,
31
+ "vocab_size": 115
32
+ }
stela/EN/01/00/orthographic_gpt2/generation_config.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 50256,
4
+ "eos_token_id": 50256,
5
+ "transformers_version": "4.49.0"
6
+ }
stela/EN/01/00/orthographic_gpt2/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c375a480783b7e3472c71b82bcd2c2236605a1903e5ba93c359adccb6c6e88e5
3
+ size 343737984
stela/EN/01/00/orthographic_gpt2/training.logs ADDED
@@ -0,0 +1,315 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 04/30/2025 14:46:58 - lexical_benchmark.train.run_train - INFO - Loading gpt2 trainer class.
2
+ 04/30/2025 14:47:03 - datasets - INFO - PyTorch version 2.5.1 available.
3
+ 04/30/2025 14:47:03 - datasets - INFO - Polars version 1.26.0 available.
4
+ 04/30/2025 14:47:03 - lexical_benchmark.train.train_params - INFO - Loading params from /lustre/fsn1/projects/rech/hhb/uye44va/workspace/lm_benchmark/prod/code/src/lexical_benchmark/train/train-params.toml
5
+ 04/30/2025 14:47:03 - lexical_benchmark.train.trainers.gpt2 - INFO - Loading char tokenizer
6
+ 04/30/2025 14:47:03 - lexical_benchmark.train.trainers.gpt2 - INFO - Tokenizer loaded with vocabulary size: 115
7
+ 04/30/2025 14:47:03 - lexical_benchmark.train.trainers.gpt2 - INFO - Loading and processing datasets with joined utterances (max_length=1024)
8
+ 04/30/2025 14:47:03 - lexical_benchmark.train.tokenizers - INFO - Reading text from /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/datasets/stela/by_size/EN/01/00/train.tokenized
9
+ 04/30/2025 14:47:03 - lexical_benchmark.train.tokenizers - INFO - Read 33094 lines from file
10
+ 04/30/2025 14:47:06 - lexical_benchmark.train.tokenizers - INFO - Created 3703 joined examples from 33094 original lines
11
+ 04/30/2025 14:47:07 - lexical_benchmark.train.tokenizers - INFO - Reading text from /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/datasets/stela/by_size/EN/dev/dev.tokenized
12
+ 04/30/2025 14:47:07 - lexical_benchmark.train.tokenizers - INFO - Read 179806 lines from file
13
+ 04/30/2025 14:47:25 - lexical_benchmark.train.tokenizers - INFO - Created 18373 joined examples from 179806 original lines
14
+ 04/30/2025 14:47:28 - lexical_benchmark.train.trainers.gpt2 - INFO - Training dataset size: 3703
15
+ 04/30/2025 14:47:28 - lexical_benchmark.train.trainers.gpt2 - INFO - Validation dataset size: 18373
16
+ 04/30/2025 14:47:28 - lexical_benchmark.train.trainers.gpt2 - INFO - Creating GPT2 configuration
17
+ 04/30/2025 14:47:28 - lexical_benchmark.train.trainers.gpt2 - INFO - Creating standard Trainer
18
+ 04/30/2025 14:47:29 - lexical_benchmark.train.run_train - INFO - Running training of stela_gpt2_EN_01_00
19
+ 04/30/2025 14:51:59 - lexical_benchmark.train.run_train - INFO - Loading gpt2 trainer class.
20
+ 04/30/2025 14:52:04 - datasets - INFO - PyTorch version 2.5.1 available.
21
+ 04/30/2025 14:52:04 - datasets - INFO - Polars version 1.26.0 available.
22
+ 04/30/2025 14:52:04 - lexical_benchmark.train.train_params - INFO - Loading params from /lustre/fsn1/projects/rech/hhb/uye44va/workspace/lm_benchmark/prod/code/src/lexical_benchmark/train/train-params.toml
23
+ 04/30/2025 14:52:04 - lexical_benchmark.train.trainers.gpt2 - INFO - Loading char tokenizer
24
+ 04/30/2025 14:52:04 - lexical_benchmark.train.trainers.gpt2 - INFO - Tokenizer loaded with vocabulary size: 115
25
+ 04/30/2025 14:52:04 - lexical_benchmark.train.trainers.gpt2 - INFO - Loading and processing datasets with joined utterances (max_length=1024)
26
+ 04/30/2025 14:52:04 - lexical_benchmark.train.tokenizers - INFO - Reading text from /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/datasets/stela/by_size/EN/01/00/train.tokenized
27
+ 04/30/2025 14:52:04 - lexical_benchmark.train.tokenizers - INFO - Read 33094 lines from file
28
+ 04/30/2025 14:52:07 - lexical_benchmark.train.tokenizers - INFO - Created 3703 joined examples from 33094 original lines
29
+ 04/30/2025 14:52:08 - lexical_benchmark.train.tokenizers - INFO - Reading text from /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/datasets/stela/by_size/EN/dev/dev.tokenized
30
+ 04/30/2025 14:52:08 - lexical_benchmark.train.tokenizers - INFO - Read 179806 lines from file
31
+ 04/30/2025 14:52:26 - lexical_benchmark.train.tokenizers - INFO - Created 18373 joined examples from 179806 original lines
32
+ 04/30/2025 14:52:29 - lexical_benchmark.train.trainers.gpt2 - INFO - Training dataset size: 3703
33
+ 04/30/2025 14:52:29 - lexical_benchmark.train.trainers.gpt2 - INFO - Validation dataset size: 18373
34
+ 04/30/2025 14:52:29 - lexical_benchmark.train.trainers.gpt2 - INFO - Creating GPT2 configuration
35
+ 04/30/2025 14:52:30 - lexical_benchmark.train.trainers.gpt2 - INFO - Creating standard Trainer
36
+ 04/30/2025 14:52:30 - lexical_benchmark.train.run_train - INFO - Running training of stela_gpt2_EN_01_00
37
+ ...
38
+ 05/23/2025 09:42:37 - urllib3.connectionpool - WARNING - Retrying (Retry(total=2, connect=None, read=None, redirect=None, status=None)) after connection broken by 'NewConnectionError('<urllib3.connection.HTTPSConnection object at 0x14e25a044f50>: Failed to establish a new connection: [Errno 101] Network is unreachable')': /api/4504800232407040/envelope/
39
+ 05/23/2025 09:42:37 - urllib3.connectionpool - WARNING - Retrying (Retry(total=1, connect=None, read=None, redirect=None, status=None)) after connection broken by 'NewConnectionError('<urllib3.connection.HTTPSConnection object at 0x14e25a059110>: Failed to establish a new connection: [Errno 101] Network is unreachable')': /api/4504800232407040/envelope/
40
+ 05/23/2025 09:42:37 - urllib3.connectionpool - WARNING - Retrying (Retry(total=0, connect=None, read=None, redirect=None, status=None)) after connection broken by 'NewConnectionError('<urllib3.connection.HTTPSConnection object at 0x14e25a059950>: Failed to establish a new connection: [Errno 101] Network is unreachable')': /api/4504800232407040/envelope/
41
+ 05/23/2025 09:42:37 - urllib3.connectionpool - WARNING - Retrying (Retry(total=2, connect=None, read=None, redirect=None, status=None)) after connection broken by 'NewConnectionError('<urllib3.connection.HTTPSConnection object at 0x14e25a011f90>: Failed to establish a new connection: [Errno 101] Network is unreachable')': /api/4504800232407040/envelope/
42
+ 05/23/2025 09:42:37 - urllib3.connectionpool - WARNING - Retrying (Retry(total=1, connect=None, read=None, redirect=None, status=None)) after connection broken by 'NewConnectionError('<urllib3.connection.HTTPSConnection object at 0x14e25a044550>: Failed to establish a new connection: [Errno 101] Network is unreachable')': /api/4504800232407040/envelope/
43
+ 05/23/2025 09:42:37 - urllib3.connectionpool - WARNING - Retrying (Retry(total=0, connect=None, read=None, redirect=None, status=None)) after connection broken by 'NewConnectionError('<urllib3.connection.HTTPSConnection object at 0x14e25a05bf90>: Failed to establish a new connection: [Errno 101] Network is unreachable')': /api/4504800232407040/envelope/
44
+ 05/23/2025 10:41:44 - urllib3.connectionpool - WARNING - Retrying (Retry(total=2, connect=None, read=None, redirect=None, status=None)) after connection broken by 'NewConnectionError('<urllib3.connection.HTTPSConnection object at 0x147c826d8110>: Failed to establish a new connection: [Errno 101] Network is unreachable')': /api/4504800232407040/envelope/
45
+ 05/23/2025 10:41:44 - urllib3.connectionpool - WARNING - Retrying (Retry(total=1, connect=None, read=None, redirect=None, status=None)) after connection broken by 'NewConnectionError('<urllib3.connection.HTTPSConnection object at 0x147c826ed850>: Failed to establish a new connection: [Errno 101] Network is unreachable')': /api/4504800232407040/envelope/
46
+ 05/23/2025 10:41:44 - urllib3.connectionpool - WARNING - Retrying (Retry(total=0, connect=None, read=None, redirect=None, status=None)) after connection broken by 'NewConnectionError('<urllib3.connection.HTTPSConnection object at 0x147c826f0310>: Failed to establish a new connection: [Errno 101] Network is unreachable')': /api/4504800232407040/envelope/
47
+ 05/23/2025 10:41:44 - urllib3.connectionpool - WARNING - Retrying (Retry(total=2, connect=None, read=None, redirect=None, status=None)) after connection broken by 'NewConnectionError('<urllib3.connection.HTTPSConnection object at 0x147c826f1290>: Failed to establish a new connection: [Errno 101] Network is unreachable')': /api/4504800232407040/envelope/
48
+ 05/23/2025 10:51:13 - urllib3.connectionpool - WARNING - Retrying (Retry(total=2, connect=None, read=None, redirect=None, status=None)) after connection broken by 'NewConnectionError('<urllib3.connection.HTTPSConnection object at 0x14a6ab688150>: Failed to establish a new connection: [Errno 101] Network is unreachable')': /api/4504800232407040/envelope/
49
+ 05/23/2025 10:51:13 - urllib3.connectionpool - WARNING - Retrying (Retry(total=1, connect=None, read=None, redirect=None, status=None)) after connection broken by 'NewConnectionError('<urllib3.connection.HTTPSConnection object at 0x14a6ab6884d0>: Failed to establish a new connection: [Errno 101] Network is unreachable')': /api/4504800232407040/envelope/
50
+ 05/23/2025 10:51:13 - urllib3.connectionpool - WARNING - Retrying (Retry(total=0, connect=None, read=None, redirect=None, status=None)) after connection broken by 'NewConnectionError('<urllib3.connection.HTTPSConnection object at 0x14a6ab698610>: Failed to establish a new connection: [Errno 101] Network is unreachable')': /api/4504800232407040/envelope/
51
+ 05/23/2025 10:51:13 - urllib3.connectionpool - WARNING - Retrying (Retry(total=2, connect=None, read=None, redirect=None, status=None)) after connection broken by 'NewConnectionError('<urllib3.connection.HTTPSConnection object at 0x14a6ab699ad0>: Failed to establish a new connection: [Errno 101] Network is unreachable')': /api/4504800232407040/envelope/
52
+ 05/23/2025 10:51:13 - urllib3.connectionpool - WARNING - Retrying (Retry(total=1, connect=None, read=None, redirect=None, status=None)) after connection broken by 'NewConnectionError('<urllib3.connection.HTTPSConnection object at 0x14a6ab6b57d0>: Failed to establish a new connection: [Errno 101] Network is unreachable')': /api/4504800232407040/envelope/
53
+ 05/23/2025 10:51:13 - urllib3.connectionpool - WARNING - Retrying (Retry(total=0, connect=None, read=None, redirect=None, status=None)) after connection broken by 'NewConnectionError('<urllib3.connection.HTTPSConnection object at 0x14a6ab6b6010>: Failed to establish a new connection: [Errno 101] Network is unreachable')': /api/4504800232407040/envelope/
54
+ 05/23/2025 11:12:51 - urllib3.connectionpool - WARNING - Retrying (Retry(total=2, connect=None, read=None, redirect=None, status=None)) after connection broken by 'NewConnectionError('<urllib3.connection.HTTPSConnection object at 0x154fd242e510>: Failed to establish a new connection: [Errno 101] Network is unreachable')': /api/4504800232407040/envelope/
55
+ 05/23/2025 11:12:51 - urllib3.connectionpool - WARNING - Retrying (Retry(total=1, connect=None, read=None, redirect=None, status=None)) after connection broken by 'NewConnectionError('<urllib3.connection.HTTPSConnection object at 0x154fd2499d10>: Failed to establish a new connection: [Errno 101] Network is unreachable')': /api/4504800232407040/envelope/
56
+ 05/23/2025 11:12:51 - urllib3.connectionpool - WARNING - Retrying (Retry(total=0, connect=None, read=None, redirect=None, status=None)) after connection broken by 'NewConnectionError('<urllib3.connection.HTTPSConnection object at 0x154fd24a4b10>: Failed to establish a new connection: [Errno 101] Network is unreachable')': /api/4504800232407040/envelope/
57
+ 05/23/2025 11:12:51 - urllib3.connectionpool - WARNING - Retrying (Retry(total=2, connect=None, read=None, redirect=None, status=None)) after connection broken by 'NewConnectionError('<urllib3.connection.HTTPSConnection object at 0x154fd242f4d0>: Failed to establish a new connection: [Errno 101] Network is unreachable')': /api/4504800232407040/envelope/
58
+ 05/23/2025 11:12:51 - urllib3.connectionpool - WARNING - Retrying (Retry(total=1, connect=None, read=None, redirect=None, status=None)) after connection broken by 'NewConnectionError('<urllib3.connection.HTTPSConnection object at 0x154fd2498f90>: Failed to establish a new connection: [Errno 101] Network is unreachable')': /api/4504800232407040/envelope/
59
+ 05/23/2025 11:12:51 - urllib3.connectionpool - WARNING - Retrying (Retry(total=0, connect=None, read=None, redirect=None, status=None)) after connection broken by 'NewConnectionError('<urllib3.connection.HTTPSConnection object at 0x154fd24a5e10>: Failed to establish a new connection: [Errno 101] Network is unreachable')': /api/4504800232407040/envelope/
60
+ 05/23/2025 11:16:30 - lexical_benchmark.train.run_train - INFO - Loading gpt2 trainer class.
61
+ 05/23/2025 11:16:45 - datasets - INFO - PyTorch version 2.5.1 available.
62
+ 05/23/2025 11:16:45 - datasets - INFO - Polars version 1.26.0 available.
63
+ 05/23/2025 11:16:46 - lexical_benchmark.train.train_params - INFO - Loading params from /lustre/fsn1/projects/rech/hhb/uye44va/workspace/lm_benchmark/prod/code/src/lexical_benchmark/train/train-params.toml
64
+ 05/23/2025 11:16:46 - lexical_benchmark.train.trainers.gpt2 - INFO - Loading char tokenizer
65
+ 05/23/2025 11:16:46 - lexical_benchmark.train.trainers.gpt2 - INFO - Tokenizer loaded with vocabulary size: 115
66
+ 05/23/2025 11:16:46 - lexical_benchmark.train.trainers.gpt2 - INFO - Loading and processing datasets with joined utterances (max_length=1024)
67
+ 05/23/2025 11:16:46 - lexical_benchmark.train.tokenizers - INFO - Reading text from /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/datasets/stela/by_size/EN/01/00/train.tokenized
68
+ 05/23/2025 11:16:46 - lexical_benchmark.train.tokenizers - INFO - Read 33094 lines from file
69
+ 05/23/2025 11:16:50 - lexical_benchmark.train.tokenizers - INFO - Created 3703 joined examples from 33094 original lines
70
+ 05/23/2025 11:16:50 - lexical_benchmark.train.tokenizers - INFO - Reading text from /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/datasets/stela/by_size/EN/dev/dev.tokenized
71
+ 05/23/2025 11:16:50 - lexical_benchmark.train.tokenizers - INFO - Read 179806 lines from file
72
+ 05/23/2025 11:17:08 - lexical_benchmark.train.tokenizers - INFO - Created 18373 joined examples from 179806 original lines
73
+ 05/23/2025 11:17:11 - lexical_benchmark.train.trainers.gpt2 - INFO - Training dataset size: 3703
74
+ 05/23/2025 11:17:11 - lexical_benchmark.train.trainers.gpt2 - INFO - Validation dataset size: 18373
75
+ 05/23/2025 11:17:11 - lexical_benchmark.train.trainers.gpt2 - INFO - Creating GPT2 configuration
76
+ 05/23/2025 11:17:12 - lexical_benchmark.train.trainers.gpt2 - INFO - Creating standard Trainer
77
+ 05/23/2025 11:17:12 - lexical_benchmark.train.run_train - INFO - Running training of stela_gpt2_EN_01_00
78
+ 05/23/2025 11:29:47 - lexical_benchmark.train.run_train - INFO - Loading gpt2 trainer class.
79
+ 05/23/2025 11:30:00 - datasets - INFO - PyTorch version 2.5.1 available.
80
+ 05/23/2025 11:30:00 - datasets - INFO - Polars version 1.26.0 available.
81
+ 05/23/2025 11:30:00 - lexical_benchmark.train.train_params - INFO - Loading params from /lustre/fsn1/projects/rech/hhb/uye44va/workspace/lm_benchmark/prod/code/src/lexical_benchmark/train/train-params.toml
82
+ 05/23/2025 11:30:00 - lexical_benchmark.train.trainers.gpt2 - INFO - Loading char tokenizer
83
+ 05/23/2025 11:30:00 - lexical_benchmark.train.trainers.gpt2 - INFO - Tokenizer loaded with vocabulary size: 115
84
+ 05/23/2025 11:30:00 - lexical_benchmark.train.trainers.gpt2 - INFO - Loading and processing datasets with joined utterances (max_length=1024)
85
+ 05/23/2025 11:30:00 - lexical_benchmark.train.tokenizers - INFO - Reading text from /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/datasets/stela/by_size/EN/01/00/train.tokenized
86
+ 05/23/2025 11:30:00 - lexical_benchmark.train.tokenizers - INFO - Read 33094 lines from file
87
+ 05/23/2025 11:30:04 - lexical_benchmark.train.tokenizers - INFO - Created 3703 joined examples from 33094 original lines
88
+ 05/23/2025 11:30:05 - lexical_benchmark.train.tokenizers - INFO - Reading text from /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/datasets/stela/by_size/EN/dev/dev.tokenized
89
+ 05/23/2025 11:30:05 - lexical_benchmark.train.tokenizers - INFO - Read 179806 lines from file
90
+ 05/23/2025 11:30:24 - lexical_benchmark.train.tokenizers - INFO - Created 18373 joined examples from 179806 original lines
91
+ 05/23/2025 11:30:28 - lexical_benchmark.train.trainers.gpt2 - INFO - Training dataset size: 3703
92
+ 05/23/2025 11:30:28 - lexical_benchmark.train.trainers.gpt2 - INFO - Validation dataset size: 18373
93
+ 05/23/2025 11:30:28 - lexical_benchmark.train.trainers.gpt2 - INFO - Creating GPT2 configuration
94
+ 05/23/2025 11:30:29 - lexical_benchmark.train.trainers.gpt2 - INFO - Creating standard Trainer
95
+ 05/23/2025 11:30:29 - lexical_benchmark.train.run_train - INFO - Running training of stela_gpt2_EN_01_00
96
+ 05/23/2025 16:13:20 - lexical_benchmark.train.run_train - INFO - Loading gpt2 trainer class.
97
+ 05/23/2025 16:13:33 - datasets - INFO - PyTorch version 2.5.1 available.
98
+ 05/23/2025 16:13:33 - datasets - INFO - Polars version 1.26.0 available.
99
+ 05/23/2025 16:13:34 - lexical_benchmark.train.train_params - INFO - Loading params from /lustre/fsn1/projects/rech/hhb/uye44va/workspace/lm_benchmark/prod/code/src/lexical_benchmark/train/train-params.toml
100
+ 05/23/2025 16:13:34 - lexical_benchmark.train.trainers.gpt2 - INFO - Loading char tokenizer
101
+ 05/23/2025 16:13:34 - lexical_benchmark.train.trainers.gpt2 - INFO - Tokenizer loaded with vocabulary size: 115
102
+ 05/23/2025 16:13:34 - lexical_benchmark.train.trainers.gpt2 - INFO - Loading and processing datasets with joined utterances (max_length=1024)
103
+ 05/23/2025 16:13:34 - lexical_benchmark.train.tokenizers - INFO - Reading text from /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/datasets/stela/by_size/EN/01/00/train.tokenized
104
+ 05/23/2025 16:13:34 - lexical_benchmark.train.tokenizers - INFO - Read 33094 lines from file
105
+ 05/23/2025 16:13:37 - lexical_benchmark.train.tokenizers - INFO - Created 3703 joined examples from 33094 original lines
106
+ 05/23/2025 16:13:38 - lexical_benchmark.train.tokenizers - INFO - Reading text from /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/datasets/stela/by_size/EN/dev/dev.tokenized
107
+ 05/23/2025 16:13:38 - lexical_benchmark.train.tokenizers - INFO - Read 179806 lines from file
108
+ 05/23/2025 16:13:55 - lexical_benchmark.train.tokenizers - INFO - Created 18373 joined examples from 179806 original lines
109
+ 05/23/2025 16:13:58 - lexical_benchmark.train.trainers.gpt2 - INFO - Training dataset size: 3703
110
+ 05/23/2025 16:13:58 - lexical_benchmark.train.trainers.gpt2 - INFO - Validation dataset size: 18373
111
+ 05/23/2025 16:13:58 - lexical_benchmark.train.trainers.gpt2 - INFO - Creating GPT2 configuration
112
+ 05/23/2025 16:13:59 - lexical_benchmark.train.trainers.gpt2 - INFO - Creating standard Trainer
113
+ 05/23/2025 16:14:00 - lexical_benchmark.train.run_train - INFO - Running training of stela_gpt2_EN_01_00
114
+ 05/23/2025 16:24:47 - lexical_benchmark.train.run_train - INFO - Loading gpt2 trainer class.
115
+ 05/23/2025 16:24:58 - datasets - INFO - PyTorch version 2.5.1 available.
116
+ 05/23/2025 16:24:58 - datasets - INFO - Polars version 1.26.0 available.
117
+ 05/23/2025 16:24:59 - lexical_benchmark.train.train_params - INFO - Loading params from /lustre/fsn1/projects/rech/hhb/uye44va/workspace/lm_benchmark/prod/code/src/lexical_benchmark/train/train-params.toml
118
+ 05/23/2025 16:24:59 - lexical_benchmark.train.trainers.gpt2 - INFO - Loading char tokenizer
119
+ 05/23/2025 16:24:59 - lexical_benchmark.train.trainers.gpt2 - INFO - Tokenizer loaded with vocabulary size: 115
120
+ 05/23/2025 16:24:59 - lexical_benchmark.train.trainers.gpt2 - INFO - Loading and processing datasets with joined utterances (max_length=1024)
121
+ 05/23/2025 16:24:59 - lexical_benchmark.train.tokenizers - INFO - Reading text from /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/datasets/stela/by_size/EN/01/00/train.tokenized
122
+ 05/23/2025 16:24:59 - lexical_benchmark.train.tokenizers - INFO - Read 33094 lines from file
123
+ 05/23/2025 16:25:02 - lexical_benchmark.train.tokenizers - INFO - Created 3703 joined examples from 33094 original lines
124
+ 05/23/2025 16:25:03 - lexical_benchmark.train.tokenizers - INFO - Reading text from /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/datasets/stela/by_size/EN/dev/dev.tokenized
125
+ 05/23/2025 16:25:03 - lexical_benchmark.train.tokenizers - INFO - Read 179806 lines from file
126
+ 05/23/2025 16:25:19 - lexical_benchmark.train.tokenizers - INFO - Created 18373 joined examples from 179806 original lines
127
+ 05/23/2025 16:25:22 - lexical_benchmark.train.trainers.gpt2 - INFO - Training dataset size: 3703
128
+ 05/23/2025 16:25:22 - lexical_benchmark.train.trainers.gpt2 - INFO - Validation dataset size: 18373
129
+ 05/23/2025 16:25:22 - lexical_benchmark.train.trainers.gpt2 - INFO - Creating GPT2 configuration
130
+ 05/23/2025 16:25:23 - lexical_benchmark.train.trainers.gpt2 - INFO - Creating standard Trainer
131
+ 05/23/2025 16:25:23 - lexical_benchmark.train.run_train - INFO - Running training of stela_gpt2_EN_01_00
132
+ 05/23/2025 16:35:44 - lexical_benchmark.train.run_train - INFO - Loading gpt2 trainer class.
133
+ 05/23/2025 16:35:48 - datasets - INFO - PyTorch version 2.5.1 available.
134
+ 05/23/2025 16:35:48 - datasets - INFO - Polars version 1.26.0 available.
135
+ 05/23/2025 16:35:48 - lexical_benchmark.train.train_params - INFO - Loading params from /lustre/fsn1/projects/rech/hhb/uye44va/workspace/lm_benchmark/prod/code/src/lexical_benchmark/train/train-params.toml
136
+ 05/23/2025 16:35:48 - lexical_benchmark.train.trainers.gpt2 - INFO - Loading char tokenizer
137
+ 05/23/2025 16:35:48 - lexical_benchmark.train.trainers.gpt2 - INFO - Tokenizer loaded with vocabulary size: 115
138
+ 05/23/2025 16:35:48 - lexical_benchmark.train.trainers.gpt2 - INFO - Loading and processing datasets with joined utterances (max_length=1024)
139
+ 05/23/2025 16:35:48 - lexical_benchmark.train.tokenizers - INFO - Reading text from /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/datasets/stela/by_size/EN/01/00/train.tokenized
140
+ 05/23/2025 16:35:48 - lexical_benchmark.train.tokenizers - INFO - Read 33094 lines from file
141
+ 05/23/2025 16:35:51 - lexical_benchmark.train.tokenizers - INFO - Created 3703 joined examples from 33094 original lines
142
+ 05/23/2025 16:35:52 - lexical_benchmark.train.tokenizers - INFO - Reading text from /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/datasets/stela/by_size/EN/dev/dev.tokenized
143
+ 05/23/2025 16:35:52 - lexical_benchmark.train.tokenizers - INFO - Read 179806 lines from file
144
+ 05/23/2025 16:36:09 - lexical_benchmark.train.tokenizers - INFO - Created 18373 joined examples from 179806 original lines
145
+ 05/23/2025 16:36:12 - lexical_benchmark.train.trainers.gpt2 - INFO - Training dataset size: 3703
146
+ 05/23/2025 16:36:12 - lexical_benchmark.train.trainers.gpt2 - INFO - Validation dataset size: 18373
147
+ 05/23/2025 16:36:12 - lexical_benchmark.train.trainers.gpt2 - INFO - Creating GPT2 configuration
148
+ 05/23/2025 16:36:12 - lexical_benchmark.train.trainers.gpt2 - INFO - Creating standard Trainer
149
+ 05/23/2025 16:36:13 - lexical_benchmark.train.run_train - INFO - Running training of stela_gpt2_EN_01_00
150
+ 05/23/2025 16:37:14 - lexical_benchmark.train.run_train - INFO - Loading gpt2 trainer class.
151
+ 05/23/2025 16:37:18 - datasets - INFO - PyTorch version 2.5.1 available.
152
+ 05/23/2025 16:37:18 - datasets - INFO - Polars version 1.26.0 available.
153
+ 05/23/2025 16:37:18 - lexical_benchmark.train.train_params - INFO - Loading params from /lustre/fsn1/projects/rech/hhb/uye44va/workspace/lm_benchmark/prod/code/src/lexical_benchmark/train/train-params.toml
154
+ 05/23/2025 16:37:18 - lexical_benchmark.train.trainers.gpt2 - INFO - Loading char tokenizer
155
+ 05/23/2025 16:37:18 - lexical_benchmark.train.trainers.gpt2 - INFO - Tokenizer loaded with vocabulary size: 115
156
+ 05/23/2025 16:37:18 - lexical_benchmark.train.trainers.gpt2 - INFO - Loading and processing datasets with joined utterances (max_length=1024)
157
+ 05/23/2025 16:37:18 - lexical_benchmark.train.tokenizers - INFO - Reading text from /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/datasets/stela/by_size/EN/01/00/train.tokenized
158
+ 05/23/2025 16:37:18 - lexical_benchmark.train.tokenizers - INFO - Read 33094 lines from file
159
+ 05/23/2025 16:37:21 - lexical_benchmark.train.tokenizers - INFO - Created 3703 joined examples from 33094 original lines
160
+ 05/23/2025 16:37:22 - lexical_benchmark.train.tokenizers - INFO - Reading text from /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/datasets/stela/by_size/EN/dev/dev.tokenized
161
+ 05/23/2025 16:37:22 - lexical_benchmark.train.tokenizers - INFO - Read 179806 lines from file
162
+ 05/23/2025 16:37:38 - lexical_benchmark.train.tokenizers - INFO - Created 18373 joined examples from 179806 original lines
163
+ 05/23/2025 16:37:41 - lexical_benchmark.train.trainers.gpt2 - INFO - Training dataset size: 3703
164
+ 05/23/2025 16:37:41 - lexical_benchmark.train.trainers.gpt2 - INFO - Validation dataset size: 18373
165
+ 05/23/2025 16:37:41 - lexical_benchmark.train.trainers.gpt2 - INFO - Creating GPT2 configuration
166
+ 05/23/2025 16:37:42 - lexical_benchmark.train.trainers.gpt2 - INFO - Creating standard Trainer
167
+ 05/23/2025 16:37:42 - lexical_benchmark.train.run_train - INFO - Running training of stela_gpt2_EN_01_00
168
+ 05/23/2025 16:37:52 - lexical_benchmark.train.run_train - INFO - Loading gpt2 trainer class.
169
+ 05/23/2025 16:37:56 - datasets - INFO - PyTorch version 2.5.1 available.
170
+ 05/23/2025 16:37:56 - datasets - INFO - Polars version 1.26.0 available.
171
+ 05/23/2025 16:37:56 - lexical_benchmark.train.train_params - INFO - Loading params from /lustre/fsn1/projects/rech/hhb/uye44va/workspace/lm_benchmark/prod/code/src/lexical_benchmark/train/train-params.toml
172
+ 05/23/2025 16:37:56 - lexical_benchmark.train.trainers.gpt2 - INFO - Loading char tokenizer
173
+ 05/23/2025 16:37:56 - lexical_benchmark.train.trainers.gpt2 - INFO - Tokenizer loaded with vocabulary size: 115
174
+ 05/23/2025 16:37:56 - lexical_benchmark.train.trainers.gpt2 - INFO - Loading and processing datasets with joined utterances (max_length=1024)
175
+ 05/23/2025 16:37:56 - lexical_benchmark.train.tokenizers - INFO - Reading text from /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/datasets/stela/by_size/EN/01/00/train.tokenized
176
+ 05/23/2025 16:37:56 - lexical_benchmark.train.tokenizers - INFO - Read 33094 lines from file
177
+ 05/23/2025 16:37:59 - lexical_benchmark.train.tokenizers - INFO - Created 3703 joined examples from 33094 original lines
178
+ 05/23/2025 16:38:00 - lexical_benchmark.train.tokenizers - INFO - Reading text from /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/datasets/stela/by_size/EN/dev/dev.tokenized
179
+ 05/23/2025 16:38:00 - lexical_benchmark.train.tokenizers - INFO - Read 179806 lines from file
180
+ 05/23/2025 16:38:17 - lexical_benchmark.train.tokenizers - INFO - Created 18373 joined examples from 179806 original lines
181
+ 05/23/2025 16:38:20 - lexical_benchmark.train.trainers.gpt2 - INFO - Training dataset size: 3703
182
+ 05/23/2025 16:38:20 - lexical_benchmark.train.trainers.gpt2 - INFO - Validation dataset size: 18373
183
+ 05/23/2025 16:38:20 - lexical_benchmark.train.trainers.gpt2 - INFO - Creating GPT2 configuration
184
+ 05/23/2025 16:38:21 - lexical_benchmark.train.trainers.gpt2 - INFO - Creating standard Trainer
185
+ 05/23/2025 16:38:21 - lexical_benchmark.train.run_train - INFO - Running training of stela_gpt2_EN_01_00
186
+ 05/23/2025 16:39:06 - lexical_benchmark.train.run_train - INFO - Loading gpt2 trainer class.
187
+ 05/23/2025 16:39:10 - datasets - INFO - PyTorch version 2.5.1 available.
188
+ 05/23/2025 16:39:10 - datasets - INFO - Polars version 1.26.0 available.
189
+ 05/23/2025 16:39:10 - lexical_benchmark.train.train_params - INFO - Loading params from /lustre/fsn1/projects/rech/hhb/uye44va/workspace/lm_benchmark/prod/code/src/lexical_benchmark/train/train-params.toml
190
+ 05/23/2025 16:39:10 - lexical_benchmark.train.trainers.gpt2 - INFO - Loading char tokenizer
191
+ 05/23/2025 16:39:10 - lexical_benchmark.train.trainers.gpt2 - INFO - Tokenizer loaded with vocabulary size: 115
192
+ 05/23/2025 16:39:10 - lexical_benchmark.train.trainers.gpt2 - INFO - Loading and processing datasets with joined utterances (max_length=1024)
193
+ 05/23/2025 16:39:10 - lexical_benchmark.train.tokenizers - INFO - Reading text from /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/datasets/stela/by_size/EN/01/00/train.tokenized
194
+ 05/23/2025 16:39:10 - lexical_benchmark.train.tokenizers - INFO - Read 33094 lines from file
195
+ 05/23/2025 16:39:14 - lexical_benchmark.train.tokenizers - INFO - Created 3703 joined examples from 33094 original lines
196
+ 05/23/2025 16:39:14 - lexical_benchmark.train.tokenizers - INFO - Reading text from /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/datasets/stela/by_size/EN/dev/dev.tokenized
197
+ 05/23/2025 16:39:14 - lexical_benchmark.train.tokenizers - INFO - Read 179806 lines from file
198
+ 05/23/2025 16:39:31 - lexical_benchmark.train.tokenizers - INFO - Created 18373 joined examples from 179806 original lines
199
+ 05/23/2025 16:39:34 - lexical_benchmark.train.trainers.gpt2 - INFO - Training dataset size: 3703
200
+ 05/23/2025 16:39:34 - lexical_benchmark.train.trainers.gpt2 - INFO - Validation dataset size: 18373
201
+ 05/23/2025 16:39:34 - lexical_benchmark.train.trainers.gpt2 - INFO - Creating GPT2 configuration
202
+ 05/23/2025 16:39:35 - lexical_benchmark.train.trainers.gpt2 - INFO - Creating standard Trainer
203
+ 05/23/2025 16:39:35 - lexical_benchmark.train.run_train - INFO - Running training of stela_gpt2_EN_01_00
204
+ 05/23/2025 17:19:00 - lexical_benchmark.train.run_train - INFO - Loading gpt2 trainer class.
205
+ 05/23/2025 17:19:11 - datasets - INFO - PyTorch version 2.5.1 available.
206
+ 05/23/2025 17:19:11 - datasets - INFO - Polars version 1.26.0 available.
207
+ 05/23/2025 17:19:12 - lexical_benchmark.train.train_params - INFO - Loading params from /lustre/fsn1/projects/rech/hhb/uye44va/workspace/lm_benchmark/prod/code/src/lexical_benchmark/train/train-params.toml
208
+ 05/23/2025 17:19:12 - lexical_benchmark.train.trainers.gpt2 - INFO - Loading char tokenizer
209
+ 05/23/2025 17:19:12 - lexical_benchmark.train.trainers.gpt2 - INFO - Tokenizer loaded with vocabulary size: 115
210
+ 05/23/2025 17:19:12 - lexical_benchmark.train.trainers.gpt2 - INFO - Loading and processing datasets with joined utterances (max_length=1024)
211
+ 05/23/2025 17:19:12 - lexical_benchmark.train.tokenizers - INFO - Reading text from /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/datasets/stela/by_size/EN/01/00/train.tokenized
212
+ 05/23/2025 17:19:12 - lexical_benchmark.train.tokenizers - INFO - Read 33094 lines from file
213
+ 05/23/2025 17:19:15 - lexical_benchmark.train.tokenizers - INFO - Created 3703 joined examples from 33094 original lines
214
+ 05/23/2025 17:19:16 - lexical_benchmark.train.tokenizers - INFO - Reading text from /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/datasets/stela/by_size/EN/dev/dev.tokenized
215
+ 05/23/2025 17:19:16 - lexical_benchmark.train.tokenizers - INFO - Read 179806 lines from file
216
+ 05/23/2025 17:19:32 - lexical_benchmark.train.tokenizers - INFO - Created 18373 joined examples from 179806 original lines
217
+ 05/23/2025 17:19:35 - lexical_benchmark.train.trainers.gpt2 - INFO - Training dataset size: 3703
218
+ 05/23/2025 17:19:35 - lexical_benchmark.train.trainers.gpt2 - INFO - Validation dataset size: 18373
219
+ 05/23/2025 17:19:35 - lexical_benchmark.train.trainers.gpt2 - INFO - Creating GPT2 configuration
220
+ 05/23/2025 17:19:36 - lexical_benchmark.train.trainers.gpt2 - INFO - Creating standard Trainer
221
+ 05/23/2025 17:19:36 - lexical_benchmark.train.run_train - INFO - Running training of stela_gpt2_EN_01_00
222
+ 05/23/2025 17:26:19 - lexical_benchmark.train.run_train - INFO - Loading gpt2 trainer class.
223
+ 05/23/2025 17:26:23 - datasets - INFO - PyTorch version 2.5.1 available.
224
+ 05/23/2025 17:26:23 - datasets - INFO - Polars version 1.26.0 available.
225
+ 05/23/2025 17:26:24 - lexical_benchmark.train.train_params - INFO - Loading params from /lustre/fsn1/projects/rech/hhb/uye44va/workspace/lm_benchmark/prod/code/src/lexical_benchmark/train/train-params.toml
226
+ 05/23/2025 17:26:24 - lexical_benchmark.train.trainers.gpt2 - INFO - Loading char tokenizer
227
+ 05/23/2025 17:26:24 - lexical_benchmark.train.trainers.gpt2 - INFO - Tokenizer loaded with vocabulary size: 115
228
+ 05/23/2025 17:26:24 - lexical_benchmark.train.trainers.gpt2 - INFO - Loading and processing datasets with joined utterances (max_length=1024)
229
+ 05/23/2025 17:26:24 - lexical_benchmark.train.tokenizers - INFO - Reading text from /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/datasets/stela/by_size/EN/01/00/train.tokenized
230
+ 05/23/2025 17:26:24 - lexical_benchmark.train.tokenizers - INFO - Read 33094 lines from file
231
+ 05/23/2025 17:26:27 - lexical_benchmark.train.tokenizers - INFO - Created 3703 joined examples from 33094 original lines
232
+ 05/23/2025 17:26:27 - lexical_benchmark.train.tokenizers - INFO - Reading text from /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/datasets/stela/by_size/EN/dev/dev.tokenized
233
+ 05/23/2025 17:26:27 - lexical_benchmark.train.tokenizers - INFO - Read 179806 lines from file
234
+ 05/23/2025 17:26:44 - lexical_benchmark.train.tokenizers - INFO - Created 18373 joined examples from 179806 original lines
235
+ 05/23/2025 17:26:47 - lexical_benchmark.train.trainers.gpt2 - INFO - Training dataset size: 3703
236
+ 05/23/2025 17:26:47 - lexical_benchmark.train.trainers.gpt2 - INFO - Validation dataset size: 18373
237
+ 05/23/2025 17:26:47 - lexical_benchmark.train.trainers.gpt2 - INFO - Creating GPT2 configuration
238
+ 05/23/2025 17:26:48 - lexical_benchmark.train.trainers.gpt2 - INFO - Creating standard Trainer
239
+ 05/23/2025 17:26:48 - lexical_benchmark.train.run_train - INFO - Running training of stela_gpt2_EN_01_00
240
+ 05/23/2025 17:30:12 - lexical_benchmark.train.run_train - INFO - Loading gpt2 trainer class.
241
+ 05/23/2025 17:30:16 - datasets - INFO - PyTorch version 2.5.1 available.
242
+ 05/23/2025 17:30:16 - datasets - INFO - Polars version 1.26.0 available.
243
+ 05/23/2025 17:30:16 - lexical_benchmark.train.train_params - INFO - Loading params from /lustre/fsn1/projects/rech/hhb/uye44va/workspace/lm_benchmark/prod/code/src/lexical_benchmark/train/train-params.toml
244
+ 05/23/2025 17:30:16 - lexical_benchmark.train.trainers.gpt2 - INFO - Loading char tokenizer
245
+ 05/23/2025 17:30:16 - lexical_benchmark.train.trainers.gpt2 - INFO - Tokenizer loaded with vocabulary size: 115
246
+ 05/23/2025 17:30:16 - lexical_benchmark.train.trainers.gpt2 - INFO - Loading and processing datasets with joined utterances (max_length=1024)
247
+ 05/23/2025 17:30:16 - lexical_benchmark.train.tokenizers - INFO - Reading text from /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/datasets/stela/by_size/EN/01/00/train.tokenized
248
+ 05/23/2025 17:30:16 - lexical_benchmark.train.tokenizers - INFO - Read 33094 lines from file
249
+ 05/23/2025 17:30:19 - lexical_benchmark.train.tokenizers - INFO - Created 3703 joined examples from 33094 original lines
250
+ 05/23/2025 17:30:20 - lexical_benchmark.train.tokenizers - INFO - Reading text from /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/datasets/stela/by_size/EN/dev/dev.tokenized
251
+ 05/23/2025 17:30:20 - lexical_benchmark.train.tokenizers - INFO - Read 179806 lines from file
252
+ 05/23/2025 17:30:36 - lexical_benchmark.train.tokenizers - INFO - Created 18373 joined examples from 179806 original lines
253
+ 05/23/2025 17:30:39 - lexical_benchmark.train.trainers.gpt2 - INFO - Training dataset size: 3703
254
+ 05/23/2025 17:30:39 - lexical_benchmark.train.trainers.gpt2 - INFO - Validation dataset size: 18373
255
+ 05/23/2025 17:30:39 - lexical_benchmark.train.trainers.gpt2 - INFO - Creating GPT2 configuration
256
+ 05/23/2025 17:30:40 - lexical_benchmark.train.trainers.gpt2 - INFO - Creating standard Trainer
257
+ 05/23/2025 17:30:40 - lexical_benchmark.train.run_train - INFO - Running training of stela_gpt2_EN_01_00
258
+ 05/23/2025 17:39:06 - lexical_benchmark.train.run_train - INFO - Loading gpt2 trainer class.
259
+ 05/23/2025 17:39:18 - datasets - INFO - PyTorch version 2.5.1 available.
260
+ 05/23/2025 17:39:18 - datasets - INFO - Polars version 1.26.0 available.
261
+ 05/23/2025 17:39:18 - lexical_benchmark.train.train_params - INFO - Loading params from /lustre/fsn1/projects/rech/hhb/uye44va/workspace/lm_benchmark/prod/code/src/lexical_benchmark/train/train-params.toml
262
+ 05/23/2025 17:39:18 - lexical_benchmark.train.trainers.gpt2 - INFO - Loading char tokenizer
263
+ 05/23/2025 17:39:18 - lexical_benchmark.train.trainers.gpt2 - INFO - Tokenizer loaded with vocabulary size: 115
264
+ 05/23/2025 17:39:18 - lexical_benchmark.train.trainers.gpt2 - INFO - Loading and processing datasets with joined utterances (max_length=1024)
265
+ 05/23/2025 17:39:18 - lexical_benchmark.train.tokenizers - INFO - Reading text from /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/datasets/stela/by_size/EN/01/00/train.tokenized
266
+ 05/23/2025 17:39:19 - lexical_benchmark.train.tokenizers - INFO - Read 33094 lines from file
267
+ 05/23/2025 17:39:22 - lexical_benchmark.train.tokenizers - INFO - Created 3703 joined examples from 33094 original lines
268
+ 05/23/2025 17:39:23 - lexical_benchmark.train.tokenizers - INFO - Reading text from /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/datasets/stela/by_size/EN/dev/dev.tokenized
269
+ 05/23/2025 17:39:23 - lexical_benchmark.train.tokenizers - INFO - Read 179806 lines from file
270
+ 05/23/2025 17:39:41 - lexical_benchmark.train.tokenizers - INFO - Created 18373 joined examples from 179806 original lines
271
+ 05/23/2025 17:39:44 - lexical_benchmark.train.trainers.gpt2 - INFO - Training dataset size: 3703
272
+ 05/23/2025 17:39:44 - lexical_benchmark.train.trainers.gpt2 - INFO - Validation dataset size: 18373
273
+ 05/23/2025 17:39:44 - lexical_benchmark.train.trainers.gpt2 - INFO - Creating GPT2 configuration
274
+ 05/23/2025 17:39:45 - lexical_benchmark.train.trainers.gpt2 - INFO - Creating standard Trainer
275
+ 05/23/2025 17:39:45 - lexical_benchmark.train.run_train - INFO - Running training of stela_gpt2_EN_01_00
276
+ 05/23/2025 18:26:14 - lexical_benchmark.train.run_train - INFO - Loading gpt2 trainer class.
277
+ 05/23/2025 18:26:18 - datasets - INFO - PyTorch version 2.5.1 available.
278
+ 05/23/2025 18:26:18 - datasets - INFO - Polars version 1.26.0 available.
279
+ 05/23/2025 18:26:19 - lexical_benchmark.train.train_params - INFO - Loading params from /lustre/fsn1/projects/rech/hhb/uye44va/workspace/lm_benchmark/prod/code/src/lexical_benchmark/train/train-params.toml
280
+ 05/23/2025 18:26:19 - lexical_benchmark.train.trainers.gpt2 - INFO - Loading char tokenizer
281
+ 05/23/2025 18:26:19 - lexical_benchmark.train.trainers.gpt2 - INFO - Tokenizer loaded with vocabulary size: 115
282
+ 05/23/2025 18:26:19 - lexical_benchmark.train.trainers.gpt2 - INFO - Loading and processing datasets with joined utterances (max_length=1024)
283
+ 05/23/2025 18:26:19 - lexical_benchmark.train.tokenizers - INFO - Reading text from /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/datasets/stela/by_size/EN/01/00/train.tokenized
284
+ 05/23/2025 18:26:19 - lexical_benchmark.train.tokenizers - INFO - Read 33094 lines from file
285
+ 05/23/2025 18:26:22 - lexical_benchmark.train.tokenizers - INFO - Created 3703 joined examples from 33094 original lines
286
+ 05/23/2025 18:26:23 - lexical_benchmark.train.tokenizers - INFO - Reading text from /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/datasets/stela/by_size/EN/dev/dev.tokenized
287
+ 05/23/2025 18:26:23 - lexical_benchmark.train.tokenizers - INFO - Read 179806 lines from file
288
+ 05/23/2025 18:26:40 - lexical_benchmark.train.tokenizers - INFO - Created 18373 joined examples from 179806 original lines
289
+ 05/23/2025 18:26:43 - lexical_benchmark.train.trainers.gpt2 - INFO - Training dataset size: 3703
290
+ 05/23/2025 18:26:43 - lexical_benchmark.train.trainers.gpt2 - INFO - Validation dataset size: 18373
291
+ 05/23/2025 18:26:43 - lexical_benchmark.train.trainers.gpt2 - INFO - Creating GPT2 configuration
292
+ 05/23/2025 18:26:44 - lexical_benchmark.train.trainers.gpt2 - INFO - Creating standard Trainer
293
+ 05/23/2025 18:26:44 - lexical_benchmark.train.run_train - INFO - Resuming previous training using /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/models/stela/EN/01/00/gpt2/checkpoint-2000
294
+ 05/23/2025 18:26:44 - lexical_benchmark.train.run_train - INFO - Running training of stela_gpt2_EN_01_00
295
+ 06/03/2025 15:20:47 - lexical_benchmark.train.run_train - INFO - Loading gpt2 trainer class.
296
+ 06/03/2025 15:20:59 - datasets - INFO - PyTorch version 2.5.1 available.
297
+ 06/03/2025 15:20:59 - datasets - INFO - Polars version 1.26.0 available.
298
+ 06/03/2025 15:21:00 - lexical_benchmark.train.train_params - INFO - Loading params from /lustre/fsn1/projects/rech/hhb/uye44va/workspace/lm_benchmark/prod/code/src/lexical_benchmark/train/train-params.toml
299
+ 06/03/2025 15:21:00 - lexical_benchmark.train.trainers.gpt2 - INFO - Loading char tokenizer
300
+ 06/03/2025 15:21:00 - lexical_benchmark.train.trainers.gpt2 - INFO - Tokenizer loaded with vocabulary size: 115
301
+ 06/03/2025 15:21:00 - lexical_benchmark.train.trainers.gpt2 - INFO - Loading and processing datasets with joined utterances (max_length=1024)
302
+ 06/03/2025 15:21:00 - lexical_benchmark.train.tokenizers - INFO - Reading text from /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/datasets/stela/by_size/EN/01/00/train.tokenized
303
+ 06/03/2025 15:21:00 - lexical_benchmark.train.tokenizers - INFO - Read 33094 lines from file
304
+ 06/03/2025 15:21:03 - lexical_benchmark.train.tokenizers - INFO - Created 3703 joined examples from 33094 original lines
305
+ 06/03/2025 15:21:04 - lexical_benchmark.train.tokenizers - INFO - Reading text from /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/datasets/stela/by_size/EN/dev/dev.tokenized
306
+ 06/03/2025 15:21:04 - lexical_benchmark.train.tokenizers - INFO - Read 179806 lines from file
307
+ 06/03/2025 15:21:22 - lexical_benchmark.train.tokenizers - INFO - Created 18373 joined examples from 179806 original lines
308
+ 06/03/2025 15:21:25 - lexical_benchmark.train.trainers.gpt2 - INFO - Training dataset size: 3703
309
+ 06/03/2025 15:21:25 - lexical_benchmark.train.trainers.gpt2 - INFO - Validation dataset size: 18373
310
+ 06/03/2025 15:21:25 - lexical_benchmark.train.trainers.gpt2 - INFO - Creating GPT2 configuration
311
+ 06/03/2025 15:21:26 - lexical_benchmark.train.trainers.gpt2 - INFO - Creating standard Trainer
312
+ 06/03/2025 15:21:26 - lexical_benchmark.train.run_train - INFO - Resuming previous training using /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/models/stela/EN/01/00/gpt2/checkpoint-3000
313
+ 06/03/2025 15:21:26 - lexical_benchmark.train.run_train - INFO - Running training of stela_gpt2_EN_01_00
314
+ 06/03/2025 16:09:19 - lexical_benchmark.train.run_train - INFO - Training of stela_gpt2_EN_01_00 has completed !
315
+ 06/03/2025 16:09:19 - lexical_benchmark.train.run_train - INFO - Model saved to /lustre/fswork/projects/rech/hhb/commun/lexical-benchmark/models/stela/EN/01/00/gpt2
stela/EN/01/00/orthographic_gpt2/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ba039a45c1b986f5a68e1de38ae9d1fc13b8e9f87931dc16fbf0fc66f97b1dd2
3
+ size 5496
stela/EN/01/00/orthographic_lstm/config.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "LSTMForLanguageModeling"
4
+ ],
5
+ "dropout": 0.1,
6
+ "embedding_dim": 200,
7
+ "hidden_size": 1024,
8
+ "model_type": "lstm",
9
+ "num_layers": 3,
10
+ "torch_dtype": "float32",
11
+ "transformers_version": "4.49.0",
12
+ "vocab_size": 115
13
+ }
stela/EN/01/00/orthographic_lstm/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a51c41b77dec2037ae7b525fbf0e4fb119fe7afd596af5399b5c2d23fed19aa6
3
+ size 87825996
stela/EN/01/00/orthographic_lstm/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f6334b0907ceada821514896ee444343648d866acb2aca043f45904a5f2640c0
3
+ size 5432