Upload folder using huggingface_hub
Browse files- all_results.json +17 -0
- checkpoint-2295/config.json +33 -0
- checkpoint-2295/model.safetensors +3 -0
- checkpoint-2295/optimizer.pt +3 -0
- checkpoint-2295/rng_state.pth +3 -0
- checkpoint-2295/scheduler.pt +3 -0
- checkpoint-2295/special_tokens_map.json +30 -0
- checkpoint-2295/tokenizer.json +0 -0
- checkpoint-2295/tokenizer_config.json +67 -0
- checkpoint-2295/trainer_state.json +125 -0
- checkpoint-2295/training_args.bin +3 -0
- config.json +33 -0
- eval_results.json +11 -0
- model.safetensors +3 -0
- predictions.txt +205 -0
- special_tokens_map.json +30 -0
- tokenizer.json +0 -0
- tokenizer_config.json +67 -0
- train_results.json +9 -0
- trainer_state.json +134 -0
- training_args.bin +3 -0
all_results.json
ADDED
|
@@ -0,0 +1,17 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"eval_accuracy": 0.7598039507865906,
|
| 4 |
+
"eval_f1": 0.8403908794788274,
|
| 5 |
+
"eval_loss": 1.1580651998519897,
|
| 6 |
+
"eval_mcc": 0.40096662922574366,
|
| 7 |
+
"eval_runtime": 0.3548,
|
| 8 |
+
"eval_samples": 204,
|
| 9 |
+
"eval_samples_per_second": 574.901,
|
| 10 |
+
"eval_steps_per_second": 73.272,
|
| 11 |
+
"total_flos": 1082018524016640.0,
|
| 12 |
+
"train_loss": 0.3891895601692283,
|
| 13 |
+
"train_runtime": 145.114,
|
| 14 |
+
"train_samples": 3668,
|
| 15 |
+
"train_samples_per_second": 126.383,
|
| 16 |
+
"train_steps_per_second": 15.815
|
| 17 |
+
}
|
checkpoint-2295/config.json
ADDED
|
@@ -0,0 +1,33 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "/home/ubuntu/utah/babylm-24/data/training/models/10M_babylm_ascii/SPM-Unigram_6144/DebertaV2-Base-10M_babylm-A",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"DebertaV2ForSequenceClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"hidden_act": "gelu",
|
| 8 |
+
"hidden_dropout_prob": 0.1,
|
| 9 |
+
"hidden_size": 768,
|
| 10 |
+
"initializer_range": 0.02,
|
| 11 |
+
"intermediate_size": 1536,
|
| 12 |
+
"label2id": {
|
| 13 |
+
"0": 0,
|
| 14 |
+
"1": 1
|
| 15 |
+
},
|
| 16 |
+
"layer_norm_eps": 1e-07,
|
| 17 |
+
"max_position_embeddings": 512,
|
| 18 |
+
"max_relative_positions": -1,
|
| 19 |
+
"model_type": "deberta-v2",
|
| 20 |
+
"num_attention_heads": 12,
|
| 21 |
+
"num_hidden_layers": 8,
|
| 22 |
+
"pad_token_id": 3,
|
| 23 |
+
"pooler_dropout": 0,
|
| 24 |
+
"pooler_hidden_act": "gelu",
|
| 25 |
+
"pooler_hidden_size": 768,
|
| 26 |
+
"pos_att_type": null,
|
| 27 |
+
"position_biased_input": true,
|
| 28 |
+
"relative_attention": false,
|
| 29 |
+
"torch_dtype": "float32",
|
| 30 |
+
"transformers_version": "4.44.2",
|
| 31 |
+
"type_vocab_size": 0,
|
| 32 |
+
"vocab_size": 6144
|
| 33 |
+
}
|
checkpoint-2295/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2086746b661a131e39e6951aacc1e99f04aa61166fafdfb875e1bc1909d99044
|
| 3 |
+
size 174103504
|
checkpoint-2295/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:38688b91bcd26286ad5044e60dcca6a43225c94b4b1f02dcfe43f6bca44f19e5
|
| 3 |
+
size 348288250
|
checkpoint-2295/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:24bcc026cb3452a2c941d9bc23542b69acc5a6a3355de450ff213ec3cf4a3c7d
|
| 3 |
+
size 14244
|
checkpoint-2295/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:eed6eb9138c85a080ac5a2612f6b35997bd0454a5909fd854316f550db9b7e74
|
| 3 |
+
size 1064
|
checkpoint-2295/special_tokens_map.json
ADDED
|
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"cls_token": {
|
| 3 |
+
"content": "[CLS]",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": false,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"mask_token": {
|
| 10 |
+
"content": "[MASK]",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"pad_token": {
|
| 17 |
+
"content": "[PAD]",
|
| 18 |
+
"lstrip": false,
|
| 19 |
+
"normalized": false,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"single_word": false
|
| 22 |
+
},
|
| 23 |
+
"sep_token": {
|
| 24 |
+
"content": "[SEP]",
|
| 25 |
+
"lstrip": false,
|
| 26 |
+
"normalized": false,
|
| 27 |
+
"rstrip": false,
|
| 28 |
+
"single_word": false
|
| 29 |
+
}
|
| 30 |
+
}
|
checkpoint-2295/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-2295/tokenizer_config.json
ADDED
|
@@ -0,0 +1,67 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "[UNK]",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"1": {
|
| 12 |
+
"content": "[CLS]",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"2": {
|
| 20 |
+
"content": "[SEP]",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
},
|
| 27 |
+
"3": {
|
| 28 |
+
"content": "[PAD]",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": true
|
| 34 |
+
},
|
| 35 |
+
"4": {
|
| 36 |
+
"content": "[MASK]",
|
| 37 |
+
"lstrip": false,
|
| 38 |
+
"normalized": false,
|
| 39 |
+
"rstrip": false,
|
| 40 |
+
"single_word": false,
|
| 41 |
+
"special": true
|
| 42 |
+
},
|
| 43 |
+
"5": {
|
| 44 |
+
"content": "[PAR]",
|
| 45 |
+
"lstrip": false,
|
| 46 |
+
"normalized": false,
|
| 47 |
+
"rstrip": false,
|
| 48 |
+
"single_word": false,
|
| 49 |
+
"special": true
|
| 50 |
+
},
|
| 51 |
+
"6": {
|
| 52 |
+
"content": "[TAB]",
|
| 53 |
+
"lstrip": false,
|
| 54 |
+
"normalized": false,
|
| 55 |
+
"rstrip": false,
|
| 56 |
+
"single_word": false,
|
| 57 |
+
"special": true
|
| 58 |
+
}
|
| 59 |
+
},
|
| 60 |
+
"clean_up_tokenization_spaces": false,
|
| 61 |
+
"cls_token": "[CLS]",
|
| 62 |
+
"mask_token": "[MASK]",
|
| 63 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 64 |
+
"pad_token": "[PAD]",
|
| 65 |
+
"sep_token": "[SEP]",
|
| 66 |
+
"tokenizer_class": "PreTrainedTokenizerFast"
|
| 67 |
+
}
|
checkpoint-2295/trainer_state.json
ADDED
|
@@ -0,0 +1,125 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.8403908794788274,
|
| 3 |
+
"best_model_checkpoint": "/home/ubuntu/utah/babylm-24/src/evaluation/results/finetune/DebertaV2-Base-10M_babylm-A/mrpc/checkpoint-2295",
|
| 4 |
+
"epoch": 5.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 2295,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 1.0,
|
| 13 |
+
"eval_accuracy": 0.6960784196853638,
|
| 14 |
+
"eval_f1": 0.8109756097560976,
|
| 15 |
+
"eval_loss": 0.5777735710144043,
|
| 16 |
+
"eval_mcc": 0.17012280758673753,
|
| 17 |
+
"eval_runtime": 0.347,
|
| 18 |
+
"eval_samples_per_second": 587.838,
|
| 19 |
+
"eval_steps_per_second": 74.921,
|
| 20 |
+
"step": 459
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 1.0893246187363834,
|
| 24 |
+
"grad_norm": 2.380143880844116,
|
| 25 |
+
"learning_rate": 1.5642701525054468e-05,
|
| 26 |
+
"loss": 0.6139,
|
| 27 |
+
"step": 500
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 2.0,
|
| 31 |
+
"eval_accuracy": 0.720588207244873,
|
| 32 |
+
"eval_f1": 0.8235294117647058,
|
| 33 |
+
"eval_loss": 0.5808757543563843,
|
| 34 |
+
"eval_mcc": 0.26984949202435965,
|
| 35 |
+
"eval_runtime": 0.3355,
|
| 36 |
+
"eval_samples_per_second": 608.081,
|
| 37 |
+
"eval_steps_per_second": 77.5,
|
| 38 |
+
"step": 918
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 2.178649237472767,
|
| 42 |
+
"grad_norm": 6.764563083648682,
|
| 43 |
+
"learning_rate": 1.1285403050108935e-05,
|
| 44 |
+
"loss": 0.5239,
|
| 45 |
+
"step": 1000
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 3.0,
|
| 49 |
+
"eval_accuracy": 0.7598039507865906,
|
| 50 |
+
"eval_f1": 0.8382838283828383,
|
| 51 |
+
"eval_loss": 0.6974886655807495,
|
| 52 |
+
"eval_mcc": 0.4042253913596411,
|
| 53 |
+
"eval_runtime": 0.3292,
|
| 54 |
+
"eval_samples_per_second": 619.643,
|
| 55 |
+
"eval_steps_per_second": 78.974,
|
| 56 |
+
"step": 1377
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 3.2679738562091503,
|
| 60 |
+
"grad_norm": 30.021303176879883,
|
| 61 |
+
"learning_rate": 6.928104575163399e-06,
|
| 62 |
+
"loss": 0.3466,
|
| 63 |
+
"step": 1500
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 4.0,
|
| 67 |
+
"eval_accuracy": 0.7549019455909729,
|
| 68 |
+
"eval_f1": 0.8387096774193549,
|
| 69 |
+
"eval_loss": 1.0106741189956665,
|
| 70 |
+
"eval_mcc": 0.3852755834560997,
|
| 71 |
+
"eval_runtime": 0.3222,
|
| 72 |
+
"eval_samples_per_second": 633.08,
|
| 73 |
+
"eval_steps_per_second": 80.687,
|
| 74 |
+
"step": 1836
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 4.357298474945534,
|
| 78 |
+
"grad_norm": 28.98902702331543,
|
| 79 |
+
"learning_rate": 2.570806100217865e-06,
|
| 80 |
+
"loss": 0.2213,
|
| 81 |
+
"step": 2000
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 5.0,
|
| 85 |
+
"eval_accuracy": 0.7598039507865906,
|
| 86 |
+
"eval_f1": 0.8403908794788274,
|
| 87 |
+
"eval_loss": 1.1580651998519897,
|
| 88 |
+
"eval_mcc": 0.40096662922574366,
|
| 89 |
+
"eval_runtime": 0.3547,
|
| 90 |
+
"eval_samples_per_second": 575.054,
|
| 91 |
+
"eval_steps_per_second": 73.291,
|
| 92 |
+
"step": 2295
|
| 93 |
+
}
|
| 94 |
+
],
|
| 95 |
+
"logging_steps": 500,
|
| 96 |
+
"max_steps": 2295,
|
| 97 |
+
"num_input_tokens_seen": 0,
|
| 98 |
+
"num_train_epochs": 5,
|
| 99 |
+
"save_steps": 500,
|
| 100 |
+
"stateful_callbacks": {
|
| 101 |
+
"EarlyStoppingCallback": {
|
| 102 |
+
"args": {
|
| 103 |
+
"early_stopping_patience": 3,
|
| 104 |
+
"early_stopping_threshold": 0.001
|
| 105 |
+
},
|
| 106 |
+
"attributes": {
|
| 107 |
+
"early_stopping_patience_counter": 0
|
| 108 |
+
}
|
| 109 |
+
},
|
| 110 |
+
"TrainerControl": {
|
| 111 |
+
"args": {
|
| 112 |
+
"should_epoch_stop": false,
|
| 113 |
+
"should_evaluate": false,
|
| 114 |
+
"should_log": false,
|
| 115 |
+
"should_save": true,
|
| 116 |
+
"should_training_stop": true
|
| 117 |
+
},
|
| 118 |
+
"attributes": {}
|
| 119 |
+
}
|
| 120 |
+
},
|
| 121 |
+
"total_flos": 1082018524016640.0,
|
| 122 |
+
"train_batch_size": 8,
|
| 123 |
+
"trial_name": null,
|
| 124 |
+
"trial_params": null
|
| 125 |
+
}
|
checkpoint-2295/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1ef98e8ba9bb3c1de25599b0c0ba0b280afed542abc00ad571a5363e04ceb366
|
| 3 |
+
size 5368
|
config.json
ADDED
|
@@ -0,0 +1,33 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "/home/ubuntu/utah/babylm-24/data/training/models/10M_babylm_ascii/SPM-Unigram_6144/DebertaV2-Base-10M_babylm-A",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"DebertaV2ForSequenceClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"hidden_act": "gelu",
|
| 8 |
+
"hidden_dropout_prob": 0.1,
|
| 9 |
+
"hidden_size": 768,
|
| 10 |
+
"initializer_range": 0.02,
|
| 11 |
+
"intermediate_size": 1536,
|
| 12 |
+
"label2id": {
|
| 13 |
+
"0": 0,
|
| 14 |
+
"1": 1
|
| 15 |
+
},
|
| 16 |
+
"layer_norm_eps": 1e-07,
|
| 17 |
+
"max_position_embeddings": 512,
|
| 18 |
+
"max_relative_positions": -1,
|
| 19 |
+
"model_type": "deberta-v2",
|
| 20 |
+
"num_attention_heads": 12,
|
| 21 |
+
"num_hidden_layers": 8,
|
| 22 |
+
"pad_token_id": 3,
|
| 23 |
+
"pooler_dropout": 0,
|
| 24 |
+
"pooler_hidden_act": "gelu",
|
| 25 |
+
"pooler_hidden_size": 768,
|
| 26 |
+
"pos_att_type": null,
|
| 27 |
+
"position_biased_input": true,
|
| 28 |
+
"relative_attention": false,
|
| 29 |
+
"torch_dtype": "float32",
|
| 30 |
+
"transformers_version": "4.44.2",
|
| 31 |
+
"type_vocab_size": 0,
|
| 32 |
+
"vocab_size": 6144
|
| 33 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,11 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"eval_accuracy": 0.7598039507865906,
|
| 4 |
+
"eval_f1": 0.8403908794788274,
|
| 5 |
+
"eval_loss": 1.1580651998519897,
|
| 6 |
+
"eval_mcc": 0.40096662922574366,
|
| 7 |
+
"eval_runtime": 0.3548,
|
| 8 |
+
"eval_samples": 204,
|
| 9 |
+
"eval_samples_per_second": 574.901,
|
| 10 |
+
"eval_steps_per_second": 73.272
|
| 11 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2086746b661a131e39e6951aacc1e99f04aa61166fafdfb875e1bc1909d99044
|
| 3 |
+
size 174103504
|
predictions.txt
ADDED
|
@@ -0,0 +1,205 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
index prediction
|
| 2 |
+
0 1
|
| 3 |
+
1 1
|
| 4 |
+
2 0
|
| 5 |
+
3 1
|
| 6 |
+
4 1
|
| 7 |
+
5 1
|
| 8 |
+
6 1
|
| 9 |
+
7 1
|
| 10 |
+
8 1
|
| 11 |
+
9 1
|
| 12 |
+
10 1
|
| 13 |
+
11 1
|
| 14 |
+
12 1
|
| 15 |
+
13 1
|
| 16 |
+
14 1
|
| 17 |
+
15 0
|
| 18 |
+
16 1
|
| 19 |
+
17 1
|
| 20 |
+
18 1
|
| 21 |
+
19 1
|
| 22 |
+
20 1
|
| 23 |
+
21 0
|
| 24 |
+
22 1
|
| 25 |
+
23 1
|
| 26 |
+
24 0
|
| 27 |
+
25 1
|
| 28 |
+
26 1
|
| 29 |
+
27 1
|
| 30 |
+
28 1
|
| 31 |
+
29 1
|
| 32 |
+
30 0
|
| 33 |
+
31 1
|
| 34 |
+
32 1
|
| 35 |
+
33 1
|
| 36 |
+
34 1
|
| 37 |
+
35 1
|
| 38 |
+
36 1
|
| 39 |
+
37 1
|
| 40 |
+
38 1
|
| 41 |
+
39 1
|
| 42 |
+
40 1
|
| 43 |
+
41 1
|
| 44 |
+
42 1
|
| 45 |
+
43 0
|
| 46 |
+
44 1
|
| 47 |
+
45 1
|
| 48 |
+
46 1
|
| 49 |
+
47 1
|
| 50 |
+
48 1
|
| 51 |
+
49 1
|
| 52 |
+
50 1
|
| 53 |
+
51 1
|
| 54 |
+
52 1
|
| 55 |
+
53 0
|
| 56 |
+
54 1
|
| 57 |
+
55 1
|
| 58 |
+
56 1
|
| 59 |
+
57 1
|
| 60 |
+
58 1
|
| 61 |
+
59 1
|
| 62 |
+
60 1
|
| 63 |
+
61 1
|
| 64 |
+
62 1
|
| 65 |
+
63 1
|
| 66 |
+
64 0
|
| 67 |
+
65 1
|
| 68 |
+
66 1
|
| 69 |
+
67 1
|
| 70 |
+
68 1
|
| 71 |
+
69 1
|
| 72 |
+
70 1
|
| 73 |
+
71 1
|
| 74 |
+
72 1
|
| 75 |
+
73 1
|
| 76 |
+
74 0
|
| 77 |
+
75 1
|
| 78 |
+
76 0
|
| 79 |
+
77 0
|
| 80 |
+
78 0
|
| 81 |
+
79 1
|
| 82 |
+
80 1
|
| 83 |
+
81 0
|
| 84 |
+
82 1
|
| 85 |
+
83 0
|
| 86 |
+
84 0
|
| 87 |
+
85 1
|
| 88 |
+
86 1
|
| 89 |
+
87 1
|
| 90 |
+
88 1
|
| 91 |
+
89 1
|
| 92 |
+
90 1
|
| 93 |
+
91 1
|
| 94 |
+
92 1
|
| 95 |
+
93 1
|
| 96 |
+
94 1
|
| 97 |
+
95 1
|
| 98 |
+
96 1
|
| 99 |
+
97 1
|
| 100 |
+
98 0
|
| 101 |
+
99 1
|
| 102 |
+
100 0
|
| 103 |
+
101 1
|
| 104 |
+
102 1
|
| 105 |
+
103 1
|
| 106 |
+
104 1
|
| 107 |
+
105 0
|
| 108 |
+
106 1
|
| 109 |
+
107 1
|
| 110 |
+
108 1
|
| 111 |
+
109 0
|
| 112 |
+
110 1
|
| 113 |
+
111 1
|
| 114 |
+
112 0
|
| 115 |
+
113 0
|
| 116 |
+
114 1
|
| 117 |
+
115 1
|
| 118 |
+
116 1
|
| 119 |
+
117 1
|
| 120 |
+
118 0
|
| 121 |
+
119 1
|
| 122 |
+
120 1
|
| 123 |
+
121 1
|
| 124 |
+
122 1
|
| 125 |
+
123 1
|
| 126 |
+
124 1
|
| 127 |
+
125 1
|
| 128 |
+
126 1
|
| 129 |
+
127 1
|
| 130 |
+
128 1
|
| 131 |
+
129 1
|
| 132 |
+
130 1
|
| 133 |
+
131 0
|
| 134 |
+
132 1
|
| 135 |
+
133 1
|
| 136 |
+
134 1
|
| 137 |
+
135 0
|
| 138 |
+
136 0
|
| 139 |
+
137 1
|
| 140 |
+
138 1
|
| 141 |
+
139 1
|
| 142 |
+
140 1
|
| 143 |
+
141 1
|
| 144 |
+
142 1
|
| 145 |
+
143 1
|
| 146 |
+
144 1
|
| 147 |
+
145 1
|
| 148 |
+
146 1
|
| 149 |
+
147 1
|
| 150 |
+
148 1
|
| 151 |
+
149 1
|
| 152 |
+
150 1
|
| 153 |
+
151 1
|
| 154 |
+
152 0
|
| 155 |
+
153 1
|
| 156 |
+
154 1
|
| 157 |
+
155 1
|
| 158 |
+
156 1
|
| 159 |
+
157 1
|
| 160 |
+
158 1
|
| 161 |
+
159 1
|
| 162 |
+
160 1
|
| 163 |
+
161 1
|
| 164 |
+
162 0
|
| 165 |
+
163 1
|
| 166 |
+
164 1
|
| 167 |
+
165 0
|
| 168 |
+
166 1
|
| 169 |
+
167 1
|
| 170 |
+
168 1
|
| 171 |
+
169 1
|
| 172 |
+
170 1
|
| 173 |
+
171 0
|
| 174 |
+
172 1
|
| 175 |
+
173 1
|
| 176 |
+
174 0
|
| 177 |
+
175 1
|
| 178 |
+
176 1
|
| 179 |
+
177 1
|
| 180 |
+
178 0
|
| 181 |
+
179 1
|
| 182 |
+
180 1
|
| 183 |
+
181 1
|
| 184 |
+
182 1
|
| 185 |
+
183 1
|
| 186 |
+
184 0
|
| 187 |
+
185 1
|
| 188 |
+
186 1
|
| 189 |
+
187 1
|
| 190 |
+
188 0
|
| 191 |
+
189 1
|
| 192 |
+
190 0
|
| 193 |
+
191 1
|
| 194 |
+
192 1
|
| 195 |
+
193 1
|
| 196 |
+
194 0
|
| 197 |
+
195 1
|
| 198 |
+
196 1
|
| 199 |
+
197 1
|
| 200 |
+
198 0
|
| 201 |
+
199 1
|
| 202 |
+
200 1
|
| 203 |
+
201 1
|
| 204 |
+
202 1
|
| 205 |
+
203 1
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"cls_token": {
|
| 3 |
+
"content": "[CLS]",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": false,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"mask_token": {
|
| 10 |
+
"content": "[MASK]",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"pad_token": {
|
| 17 |
+
"content": "[PAD]",
|
| 18 |
+
"lstrip": false,
|
| 19 |
+
"normalized": false,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"single_word": false
|
| 22 |
+
},
|
| 23 |
+
"sep_token": {
|
| 24 |
+
"content": "[SEP]",
|
| 25 |
+
"lstrip": false,
|
| 26 |
+
"normalized": false,
|
| 27 |
+
"rstrip": false,
|
| 28 |
+
"single_word": false
|
| 29 |
+
}
|
| 30 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,67 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "[UNK]",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"1": {
|
| 12 |
+
"content": "[CLS]",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"2": {
|
| 20 |
+
"content": "[SEP]",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
},
|
| 27 |
+
"3": {
|
| 28 |
+
"content": "[PAD]",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": true
|
| 34 |
+
},
|
| 35 |
+
"4": {
|
| 36 |
+
"content": "[MASK]",
|
| 37 |
+
"lstrip": false,
|
| 38 |
+
"normalized": false,
|
| 39 |
+
"rstrip": false,
|
| 40 |
+
"single_word": false,
|
| 41 |
+
"special": true
|
| 42 |
+
},
|
| 43 |
+
"5": {
|
| 44 |
+
"content": "[PAR]",
|
| 45 |
+
"lstrip": false,
|
| 46 |
+
"normalized": false,
|
| 47 |
+
"rstrip": false,
|
| 48 |
+
"single_word": false,
|
| 49 |
+
"special": true
|
| 50 |
+
},
|
| 51 |
+
"6": {
|
| 52 |
+
"content": "[TAB]",
|
| 53 |
+
"lstrip": false,
|
| 54 |
+
"normalized": false,
|
| 55 |
+
"rstrip": false,
|
| 56 |
+
"single_word": false,
|
| 57 |
+
"special": true
|
| 58 |
+
}
|
| 59 |
+
},
|
| 60 |
+
"clean_up_tokenization_spaces": false,
|
| 61 |
+
"cls_token": "[CLS]",
|
| 62 |
+
"mask_token": "[MASK]",
|
| 63 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 64 |
+
"pad_token": "[PAD]",
|
| 65 |
+
"sep_token": "[SEP]",
|
| 66 |
+
"tokenizer_class": "PreTrainedTokenizerFast"
|
| 67 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"total_flos": 1082018524016640.0,
|
| 4 |
+
"train_loss": 0.3891895601692283,
|
| 5 |
+
"train_runtime": 145.114,
|
| 6 |
+
"train_samples": 3668,
|
| 7 |
+
"train_samples_per_second": 126.383,
|
| 8 |
+
"train_steps_per_second": 15.815
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,134 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.8403908794788274,
|
| 3 |
+
"best_model_checkpoint": "/home/ubuntu/utah/babylm-24/src/evaluation/results/finetune/DebertaV2-Base-10M_babylm-A/mrpc/checkpoint-2295",
|
| 4 |
+
"epoch": 5.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 2295,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 1.0,
|
| 13 |
+
"eval_accuracy": 0.6960784196853638,
|
| 14 |
+
"eval_f1": 0.8109756097560976,
|
| 15 |
+
"eval_loss": 0.5777735710144043,
|
| 16 |
+
"eval_mcc": 0.17012280758673753,
|
| 17 |
+
"eval_runtime": 0.347,
|
| 18 |
+
"eval_samples_per_second": 587.838,
|
| 19 |
+
"eval_steps_per_second": 74.921,
|
| 20 |
+
"step": 459
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 1.0893246187363834,
|
| 24 |
+
"grad_norm": 2.380143880844116,
|
| 25 |
+
"learning_rate": 1.5642701525054468e-05,
|
| 26 |
+
"loss": 0.6139,
|
| 27 |
+
"step": 500
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 2.0,
|
| 31 |
+
"eval_accuracy": 0.720588207244873,
|
| 32 |
+
"eval_f1": 0.8235294117647058,
|
| 33 |
+
"eval_loss": 0.5808757543563843,
|
| 34 |
+
"eval_mcc": 0.26984949202435965,
|
| 35 |
+
"eval_runtime": 0.3355,
|
| 36 |
+
"eval_samples_per_second": 608.081,
|
| 37 |
+
"eval_steps_per_second": 77.5,
|
| 38 |
+
"step": 918
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 2.178649237472767,
|
| 42 |
+
"grad_norm": 6.764563083648682,
|
| 43 |
+
"learning_rate": 1.1285403050108935e-05,
|
| 44 |
+
"loss": 0.5239,
|
| 45 |
+
"step": 1000
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 3.0,
|
| 49 |
+
"eval_accuracy": 0.7598039507865906,
|
| 50 |
+
"eval_f1": 0.8382838283828383,
|
| 51 |
+
"eval_loss": 0.6974886655807495,
|
| 52 |
+
"eval_mcc": 0.4042253913596411,
|
| 53 |
+
"eval_runtime": 0.3292,
|
| 54 |
+
"eval_samples_per_second": 619.643,
|
| 55 |
+
"eval_steps_per_second": 78.974,
|
| 56 |
+
"step": 1377
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 3.2679738562091503,
|
| 60 |
+
"grad_norm": 30.021303176879883,
|
| 61 |
+
"learning_rate": 6.928104575163399e-06,
|
| 62 |
+
"loss": 0.3466,
|
| 63 |
+
"step": 1500
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 4.0,
|
| 67 |
+
"eval_accuracy": 0.7549019455909729,
|
| 68 |
+
"eval_f1": 0.8387096774193549,
|
| 69 |
+
"eval_loss": 1.0106741189956665,
|
| 70 |
+
"eval_mcc": 0.3852755834560997,
|
| 71 |
+
"eval_runtime": 0.3222,
|
| 72 |
+
"eval_samples_per_second": 633.08,
|
| 73 |
+
"eval_steps_per_second": 80.687,
|
| 74 |
+
"step": 1836
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 4.357298474945534,
|
| 78 |
+
"grad_norm": 28.98902702331543,
|
| 79 |
+
"learning_rate": 2.570806100217865e-06,
|
| 80 |
+
"loss": 0.2213,
|
| 81 |
+
"step": 2000
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 5.0,
|
| 85 |
+
"eval_accuracy": 0.7598039507865906,
|
| 86 |
+
"eval_f1": 0.8403908794788274,
|
| 87 |
+
"eval_loss": 1.1580651998519897,
|
| 88 |
+
"eval_mcc": 0.40096662922574366,
|
| 89 |
+
"eval_runtime": 0.3547,
|
| 90 |
+
"eval_samples_per_second": 575.054,
|
| 91 |
+
"eval_steps_per_second": 73.291,
|
| 92 |
+
"step": 2295
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 5.0,
|
| 96 |
+
"step": 2295,
|
| 97 |
+
"total_flos": 1082018524016640.0,
|
| 98 |
+
"train_loss": 0.3891895601692283,
|
| 99 |
+
"train_runtime": 145.114,
|
| 100 |
+
"train_samples_per_second": 126.383,
|
| 101 |
+
"train_steps_per_second": 15.815
|
| 102 |
+
}
|
| 103 |
+
],
|
| 104 |
+
"logging_steps": 500,
|
| 105 |
+
"max_steps": 2295,
|
| 106 |
+
"num_input_tokens_seen": 0,
|
| 107 |
+
"num_train_epochs": 5,
|
| 108 |
+
"save_steps": 500,
|
| 109 |
+
"stateful_callbacks": {
|
| 110 |
+
"EarlyStoppingCallback": {
|
| 111 |
+
"args": {
|
| 112 |
+
"early_stopping_patience": 3,
|
| 113 |
+
"early_stopping_threshold": 0.001
|
| 114 |
+
},
|
| 115 |
+
"attributes": {
|
| 116 |
+
"early_stopping_patience_counter": 0
|
| 117 |
+
}
|
| 118 |
+
},
|
| 119 |
+
"TrainerControl": {
|
| 120 |
+
"args": {
|
| 121 |
+
"should_epoch_stop": false,
|
| 122 |
+
"should_evaluate": false,
|
| 123 |
+
"should_log": false,
|
| 124 |
+
"should_save": true,
|
| 125 |
+
"should_training_stop": true
|
| 126 |
+
},
|
| 127 |
+
"attributes": {}
|
| 128 |
+
}
|
| 129 |
+
},
|
| 130 |
+
"total_flos": 1082018524016640.0,
|
| 131 |
+
"train_batch_size": 8,
|
| 132 |
+
"trial_name": null,
|
| 133 |
+
"trial_params": null
|
| 134 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1ef98e8ba9bb3c1de25599b0c0ba0b280afed542abc00ad571a5363e04ceb366
|
| 3 |
+
size 5368
|