Add nboost-pt-biobert-base-msmarco-E2-S1-Mpairwise-FullDataTrue
Browse filesUpload fine-tuned rerankers for BioASQ 14B
Co-authored-by: André Ribeiro <andrepedro2004@hotmail.com>
Co-authored-by: Rúben Garrido <rubengarrido@ua.pt>
- nboost-pt-biobert-base-msmarco-E2-S1-Mpairwise-FullDataTrue/config.json +36 -0
- nboost-pt-biobert-base-msmarco-E2-S1-Mpairwise-FullDataTrue/model.safetensors +3 -0
- nboost-pt-biobert-base-msmarco-E2-S1-Mpairwise-FullDataTrue/ranx_results.json +49 -0
- nboost-pt-biobert-base-msmarco-E2-S1-Mpairwise-FullDataTrue/tokenizer.json +0 -0
- nboost-pt-biobert-base-msmarco-E2-S1-Mpairwise-FullDataTrue/tokenizer_config.json +14 -0
- nboost-pt-biobert-base-msmarco-E2-S1-Mpairwise-FullDataTrue/trainer_state.json +673 -0
- nboost-pt-biobert-base-msmarco-E2-S1-Mpairwise-FullDataTrue/training_args.bin +3 -0
nboost-pt-biobert-base-msmarco-E2-S1-Mpairwise-FullDataTrue/config.json
ADDED
|
@@ -0,0 +1,36 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_cross_attention": false,
|
| 3 |
+
"architectures": [
|
| 4 |
+
"BertForSequenceClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"bos_token_id": null,
|
| 8 |
+
"classifier_dropout": null,
|
| 9 |
+
"dtype": "bfloat16",
|
| 10 |
+
"eos_token_id": null,
|
| 11 |
+
"eos_token_ids": null,
|
| 12 |
+
"hidden_act": "gelu",
|
| 13 |
+
"hidden_dropout_prob": 0.1,
|
| 14 |
+
"hidden_size": 768,
|
| 15 |
+
"id2label": {
|
| 16 |
+
"0": "LABEL_0"
|
| 17 |
+
},
|
| 18 |
+
"initializer_range": 0.02,
|
| 19 |
+
"intermediate_size": 3072,
|
| 20 |
+
"is_decoder": false,
|
| 21 |
+
"label2id": {
|
| 22 |
+
"LABEL_0": 0
|
| 23 |
+
},
|
| 24 |
+
"layer_norm_eps": 1e-12,
|
| 25 |
+
"max_position_embeddings": 512,
|
| 26 |
+
"model_type": "bert",
|
| 27 |
+
"num_attention_heads": 12,
|
| 28 |
+
"num_hidden_layers": 12,
|
| 29 |
+
"output_past": true,
|
| 30 |
+
"pad_token_id": 0,
|
| 31 |
+
"tie_word_embeddings": true,
|
| 32 |
+
"transformers_version": "5.2.0",
|
| 33 |
+
"type_vocab_size": 2,
|
| 34 |
+
"use_cache": false,
|
| 35 |
+
"vocab_size": 28996
|
| 36 |
+
}
|
nboost-pt-biobert-base-msmarco-E2-S1-Mpairwise-FullDataTrue/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c15bc47c64c64505077e864af92a515eaee94edec6e4e9eeaa3b9054ebe56b3f
|
| 3 |
+
size 216645674
|
nboost-pt-biobert-base-msmarco-E2-S1-Mpairwise-FullDataTrue/ranx_results.json
ADDED
|
@@ -0,0 +1,49 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"model": "nboost-pt-biobert-base-msmarco",
|
| 3 |
+
"val_files": [
|
| 4 |
+
"../../data/val_data/13B3_golden.json",
|
| 5 |
+
"../../data/val_data/13B1_golden.json",
|
| 6 |
+
"../../data/val_data/13B2_golden.json",
|
| 7 |
+
"../../data/val_data/13B4_golden.json"
|
| 8 |
+
],
|
| 9 |
+
"total": {
|
| 10 |
+
"ndcg@5": 0.747188091050092,
|
| 11 |
+
"mrr": 0.864354681266446,
|
| 12 |
+
"recall@10": 0.43244704854682187,
|
| 13 |
+
"recall@100": 0.948380268525564,
|
| 14 |
+
"map@10": 0.37738149834211293,
|
| 15 |
+
"map-bioasq@10": 0.6274464674758793
|
| 16 |
+
},
|
| 17 |
+
"13B3_golden.json": {
|
| 18 |
+
"ndcg@5": 0.7365795228424811,
|
| 19 |
+
"mrr": 0.8772128851540615,
|
| 20 |
+
"recall@10": 0.43077709022924676,
|
| 21 |
+
"recall@100": 0.9612005765566681,
|
| 22 |
+
"map@10": 0.383671961401949,
|
| 23 |
+
"map-bioasq@10": 0.611980651519867
|
| 24 |
+
},
|
| 25 |
+
"13B1_golden.json": {
|
| 26 |
+
"ndcg@5": 0.7083921622159625,
|
| 27 |
+
"mrr": 0.807408963585434,
|
| 28 |
+
"recall@10": 0.4986321959565673,
|
| 29 |
+
"recall@100": 0.9629103836943025,
|
| 30 |
+
"map@10": 0.4186991399959262,
|
| 31 |
+
"map-bioasq@10": 0.5794541964934122
|
| 32 |
+
},
|
| 33 |
+
"13B2_golden.json": {
|
| 34 |
+
"ndcg@5": 0.8258869635750175,
|
| 35 |
+
"mrr": 0.9128342245989305,
|
| 36 |
+
"recall@10": 0.5147484637790445,
|
| 37 |
+
"recall@100": 0.9482215088457647,
|
| 38 |
+
"map@10": 0.47914793976451814,
|
| 39 |
+
"map-bioasq@10": 0.7324179375453884
|
| 40 |
+
},
|
| 41 |
+
"13B4_golden.json": {
|
| 42 |
+
"ndcg@5": 0.717893715566907,
|
| 43 |
+
"mrr": 0.8599626517273574,
|
| 44 |
+
"recall@10": 0.28563044422242917,
|
| 45 |
+
"recall@100": 0.9211886050055206,
|
| 46 |
+
"map@10": 0.22800695220605846,
|
| 47 |
+
"map-bioasq@10": 0.585933084344849
|
| 48 |
+
}
|
| 49 |
+
}
|
nboost-pt-biobert-base-msmarco-E2-S1-Mpairwise-FullDataTrue/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
nboost-pt-biobert-base-msmarco-E2-S1-Mpairwise-FullDataTrue/tokenizer_config.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"backend": "tokenizers",
|
| 3 |
+
"cls_token": "[CLS]",
|
| 4 |
+
"do_lower_case": false,
|
| 5 |
+
"is_local": false,
|
| 6 |
+
"mask_token": "[MASK]",
|
| 7 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 8 |
+
"pad_token": "[PAD]",
|
| 9 |
+
"sep_token": "[SEP]",
|
| 10 |
+
"strip_accents": null,
|
| 11 |
+
"tokenize_chinese_chars": true,
|
| 12 |
+
"tokenizer_class": "BertTokenizer",
|
| 13 |
+
"unk_token": "[UNK]"
|
| 14 |
+
}
|
nboost-pt-biobert-base-msmarco-E2-S1-Mpairwise-FullDataTrue/trainer_state.json
ADDED
|
@@ -0,0 +1,673 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 2.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 4452,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.022461814914645103,
|
| 14 |
+
"grad_norm": 10.375,
|
| 15 |
+
"learning_rate": 1.977987421383648e-05,
|
| 16 |
+
"loss": 0.35307395935058594,
|
| 17 |
+
"step": 50
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.044923629829290206,
|
| 21 |
+
"grad_norm": 7.3125,
|
| 22 |
+
"learning_rate": 1.955525606469003e-05,
|
| 23 |
+
"loss": 0.35419837951660155,
|
| 24 |
+
"step": 100
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.0673854447439353,
|
| 28 |
+
"grad_norm": 7.53125,
|
| 29 |
+
"learning_rate": 1.9330637915543576e-05,
|
| 30 |
+
"loss": 0.3191358947753906,
|
| 31 |
+
"step": 150
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.08984725965858041,
|
| 35 |
+
"grad_norm": 9.875,
|
| 36 |
+
"learning_rate": 1.9106019766397127e-05,
|
| 37 |
+
"loss": 0.33251625061035156,
|
| 38 |
+
"step": 200
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.11230907457322552,
|
| 42 |
+
"grad_norm": 8.8125,
|
| 43 |
+
"learning_rate": 1.8881401617250677e-05,
|
| 44 |
+
"loss": 0.3634188461303711,
|
| 45 |
+
"step": 250
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.1347708894878706,
|
| 49 |
+
"grad_norm": 10.125,
|
| 50 |
+
"learning_rate": 1.8656783468104224e-05,
|
| 51 |
+
"loss": 0.3543392562866211,
|
| 52 |
+
"step": 300
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.15723270440251572,
|
| 56 |
+
"grad_norm": 5.65625,
|
| 57 |
+
"learning_rate": 1.843216531895777e-05,
|
| 58 |
+
"loss": 0.35604286193847656,
|
| 59 |
+
"step": 350
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.17969451931716082,
|
| 63 |
+
"grad_norm": 10.1875,
|
| 64 |
+
"learning_rate": 1.8207547169811322e-05,
|
| 65 |
+
"loss": 0.35010604858398436,
|
| 66 |
+
"step": 400
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.20215633423180593,
|
| 70 |
+
"grad_norm": 7.75,
|
| 71 |
+
"learning_rate": 1.7982929020664873e-05,
|
| 72 |
+
"loss": 0.30715930938720704,
|
| 73 |
+
"step": 450
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.22461814914645103,
|
| 77 |
+
"grad_norm": 11.875,
|
| 78 |
+
"learning_rate": 1.775831087151842e-05,
|
| 79 |
+
"loss": 0.32746646881103514,
|
| 80 |
+
"step": 500
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.24707996406109614,
|
| 84 |
+
"grad_norm": 10.5625,
|
| 85 |
+
"learning_rate": 1.753369272237197e-05,
|
| 86 |
+
"loss": 0.2940776634216309,
|
| 87 |
+
"step": 550
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.2695417789757412,
|
| 91 |
+
"grad_norm": 9.1875,
|
| 92 |
+
"learning_rate": 1.7309074573225517e-05,
|
| 93 |
+
"loss": 0.31212696075439456,
|
| 94 |
+
"step": 600
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.2920035938903863,
|
| 98 |
+
"grad_norm": 8.625,
|
| 99 |
+
"learning_rate": 1.7084456424079068e-05,
|
| 100 |
+
"loss": 0.3173513412475586,
|
| 101 |
+
"step": 650
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.31446540880503143,
|
| 105 |
+
"grad_norm": 9.25,
|
| 106 |
+
"learning_rate": 1.6859838274932615e-05,
|
| 107 |
+
"loss": 0.3336635589599609,
|
| 108 |
+
"step": 700
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.33692722371967654,
|
| 112 |
+
"grad_norm": 7.9375,
|
| 113 |
+
"learning_rate": 1.6635220125786166e-05,
|
| 114 |
+
"loss": 0.3208817672729492,
|
| 115 |
+
"step": 750
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.35938903863432164,
|
| 119 |
+
"grad_norm": 9.875,
|
| 120 |
+
"learning_rate": 1.6410601976639713e-05,
|
| 121 |
+
"loss": 0.327133903503418,
|
| 122 |
+
"step": 800
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.38185085354896675,
|
| 126 |
+
"grad_norm": 9.6875,
|
| 127 |
+
"learning_rate": 1.6185983827493263e-05,
|
| 128 |
+
"loss": 0.3520771789550781,
|
| 129 |
+
"step": 850
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.40431266846361186,
|
| 133 |
+
"grad_norm": 8.25,
|
| 134 |
+
"learning_rate": 1.596136567834681e-05,
|
| 135 |
+
"loss": 0.2915140151977539,
|
| 136 |
+
"step": 900
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.42677448337825696,
|
| 140 |
+
"grad_norm": 11.6875,
|
| 141 |
+
"learning_rate": 1.573674752920036e-05,
|
| 142 |
+
"loss": 0.2957220268249512,
|
| 143 |
+
"step": 950
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.44923629829290207,
|
| 147 |
+
"grad_norm": 10.625,
|
| 148 |
+
"learning_rate": 1.5512129380053908e-05,
|
| 149 |
+
"loss": 0.3144234848022461,
|
| 150 |
+
"step": 1000
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.4716981132075472,
|
| 154 |
+
"grad_norm": 9.75,
|
| 155 |
+
"learning_rate": 1.528751123090746e-05,
|
| 156 |
+
"loss": 0.33309947967529296,
|
| 157 |
+
"step": 1050
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.4941599281221923,
|
| 161 |
+
"grad_norm": 7.625,
|
| 162 |
+
"learning_rate": 1.5062893081761007e-05,
|
| 163 |
+
"loss": 0.3460280609130859,
|
| 164 |
+
"step": 1100
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.5166217430368374,
|
| 168 |
+
"grad_norm": 9.75,
|
| 169 |
+
"learning_rate": 1.4838274932614556e-05,
|
| 170 |
+
"loss": 0.3476744842529297,
|
| 171 |
+
"step": 1150
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.5390835579514824,
|
| 175 |
+
"grad_norm": 14.0,
|
| 176 |
+
"learning_rate": 1.4613656783468107e-05,
|
| 177 |
+
"loss": 0.3004001808166504,
|
| 178 |
+
"step": 1200
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.5615453728661276,
|
| 182 |
+
"grad_norm": 12.125,
|
| 183 |
+
"learning_rate": 1.4389038634321656e-05,
|
| 184 |
+
"loss": 0.29221229553222655,
|
| 185 |
+
"step": 1250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.5840071877807727,
|
| 189 |
+
"grad_norm": 14.5625,
|
| 190 |
+
"learning_rate": 1.4164420485175203e-05,
|
| 191 |
+
"loss": 0.3244548034667969,
|
| 192 |
+
"step": 1300
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.6064690026954178,
|
| 196 |
+
"grad_norm": 6.5,
|
| 197 |
+
"learning_rate": 1.3939802336028752e-05,
|
| 198 |
+
"loss": 0.2775107192993164,
|
| 199 |
+
"step": 1350
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.6289308176100629,
|
| 203 |
+
"grad_norm": 8.1875,
|
| 204 |
+
"learning_rate": 1.3715184186882302e-05,
|
| 205 |
+
"loss": 0.30810970306396485,
|
| 206 |
+
"step": 1400
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.651392632524708,
|
| 210 |
+
"grad_norm": 15.625,
|
| 211 |
+
"learning_rate": 1.3490566037735851e-05,
|
| 212 |
+
"loss": 0.3027694320678711,
|
| 213 |
+
"step": 1450
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.6738544474393531,
|
| 217 |
+
"grad_norm": 12.3125,
|
| 218 |
+
"learning_rate": 1.3265947888589398e-05,
|
| 219 |
+
"loss": 0.3071010208129883,
|
| 220 |
+
"step": 1500
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.6963162623539982,
|
| 224 |
+
"grad_norm": 7.15625,
|
| 225 |
+
"learning_rate": 1.3041329739442947e-05,
|
| 226 |
+
"loss": 0.3080667304992676,
|
| 227 |
+
"step": 1550
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.7187780772686433,
|
| 231 |
+
"grad_norm": 8.125,
|
| 232 |
+
"learning_rate": 1.2816711590296497e-05,
|
| 233 |
+
"loss": 0.31681983947753906,
|
| 234 |
+
"step": 1600
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.7412398921832885,
|
| 238 |
+
"grad_norm": 8.75,
|
| 239 |
+
"learning_rate": 1.2592093441150046e-05,
|
| 240 |
+
"loss": 0.34106666564941407,
|
| 241 |
+
"step": 1650
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.7637017070979335,
|
| 245 |
+
"grad_norm": 16.625,
|
| 246 |
+
"learning_rate": 1.2367475292003595e-05,
|
| 247 |
+
"loss": 0.32158096313476564,
|
| 248 |
+
"step": 1700
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.7861635220125787,
|
| 252 |
+
"grad_norm": 11.875,
|
| 253 |
+
"learning_rate": 1.2142857142857142e-05,
|
| 254 |
+
"loss": 0.3163252258300781,
|
| 255 |
+
"step": 1750
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.8086253369272237,
|
| 259 |
+
"grad_norm": 10.25,
|
| 260 |
+
"learning_rate": 1.1918238993710693e-05,
|
| 261 |
+
"loss": 0.29589111328125,
|
| 262 |
+
"step": 1800
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.8310871518418689,
|
| 266 |
+
"grad_norm": 15.5625,
|
| 267 |
+
"learning_rate": 1.1693620844564242e-05,
|
| 268 |
+
"loss": 0.2735770797729492,
|
| 269 |
+
"step": 1850
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.8535489667565139,
|
| 273 |
+
"grad_norm": 11.375,
|
| 274 |
+
"learning_rate": 1.146900269541779e-05,
|
| 275 |
+
"loss": 0.2837497711181641,
|
| 276 |
+
"step": 1900
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.876010781671159,
|
| 280 |
+
"grad_norm": 13.4375,
|
| 281 |
+
"learning_rate": 1.124438454627134e-05,
|
| 282 |
+
"loss": 0.28498140335083005,
|
| 283 |
+
"step": 1950
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.8984725965858041,
|
| 287 |
+
"grad_norm": 9.1875,
|
| 288 |
+
"learning_rate": 1.101976639712489e-05,
|
| 289 |
+
"loss": 0.30365570068359377,
|
| 290 |
+
"step": 2000
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.9209344115004492,
|
| 294 |
+
"grad_norm": 15.5625,
|
| 295 |
+
"learning_rate": 1.0795148247978437e-05,
|
| 296 |
+
"loss": 0.29460960388183594,
|
| 297 |
+
"step": 2050
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.9433962264150944,
|
| 301 |
+
"grad_norm": 17.625,
|
| 302 |
+
"learning_rate": 1.0570530098831986e-05,
|
| 303 |
+
"loss": 0.33808418273925783,
|
| 304 |
+
"step": 2100
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.9658580413297394,
|
| 308 |
+
"grad_norm": 10.875,
|
| 309 |
+
"learning_rate": 1.0345911949685536e-05,
|
| 310 |
+
"loss": 0.3489389419555664,
|
| 311 |
+
"step": 2150
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.9883198562443846,
|
| 315 |
+
"grad_norm": 17.875,
|
| 316 |
+
"learning_rate": 1.0121293800539085e-05,
|
| 317 |
+
"loss": 0.34056243896484373,
|
| 318 |
+
"step": 2200
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 1.0,
|
| 322 |
+
"eval_loss": 0.3806288540363312,
|
| 323 |
+
"eval_runtime": 2.0851,
|
| 324 |
+
"eval_samples_per_second": 162.105,
|
| 325 |
+
"eval_steps_per_second": 5.276,
|
| 326 |
+
"step": 2226
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 1.0107816711590296,
|
| 330 |
+
"grad_norm": 14.25,
|
| 331 |
+
"learning_rate": 9.896675651392634e-06,
|
| 332 |
+
"loss": 0.28267147064208986,
|
| 333 |
+
"step": 2250
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 1.0332434860736748,
|
| 337 |
+
"grad_norm": 10.0,
|
| 338 |
+
"learning_rate": 9.672057502246183e-06,
|
| 339 |
+
"loss": 0.30427501678466795,
|
| 340 |
+
"step": 2300
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 1.05570530098832,
|
| 344 |
+
"grad_norm": 15.125,
|
| 345 |
+
"learning_rate": 9.447439353099732e-06,
|
| 346 |
+
"loss": 0.3022786712646484,
|
| 347 |
+
"step": 2350
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 1.0781671159029649,
|
| 351 |
+
"grad_norm": 10.8125,
|
| 352 |
+
"learning_rate": 9.22282120395328e-06,
|
| 353 |
+
"loss": 0.336663818359375,
|
| 354 |
+
"step": 2400
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 1.10062893081761,
|
| 358 |
+
"grad_norm": 13.0625,
|
| 359 |
+
"learning_rate": 8.99820305480683e-06,
|
| 360 |
+
"loss": 0.33252601623535155,
|
| 361 |
+
"step": 2450
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 1.1230907457322552,
|
| 365 |
+
"grad_norm": 10.8125,
|
| 366 |
+
"learning_rate": 8.773584905660378e-06,
|
| 367 |
+
"loss": 0.28596717834472657,
|
| 368 |
+
"step": 2500
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 1.1455525606469004,
|
| 372 |
+
"grad_norm": 5.875,
|
| 373 |
+
"learning_rate": 8.548966756513927e-06,
|
| 374 |
+
"loss": 0.29772865295410156,
|
| 375 |
+
"step": 2550
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 1.1680143755615453,
|
| 379 |
+
"grad_norm": 10.375,
|
| 380 |
+
"learning_rate": 8.324348607367476e-06,
|
| 381 |
+
"loss": 0.2924853515625,
|
| 382 |
+
"step": 2600
|
| 383 |
+
},
|
| 384 |
+
{
|
| 385 |
+
"epoch": 1.1904761904761905,
|
| 386 |
+
"grad_norm": 10.9375,
|
| 387 |
+
"learning_rate": 8.099730458221025e-06,
|
| 388 |
+
"loss": 0.3177211761474609,
|
| 389 |
+
"step": 2650
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 1.2129380053908356,
|
| 393 |
+
"grad_norm": 12.125,
|
| 394 |
+
"learning_rate": 7.875112309074573e-06,
|
| 395 |
+
"loss": 0.2946417999267578,
|
| 396 |
+
"step": 2700
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 1.2353998203054806,
|
| 400 |
+
"grad_norm": 7.625,
|
| 401 |
+
"learning_rate": 7.650494159928122e-06,
|
| 402 |
+
"loss": 0.28809988021850585,
|
| 403 |
+
"step": 2750
|
| 404 |
+
},
|
| 405 |
+
{
|
| 406 |
+
"epoch": 1.2578616352201257,
|
| 407 |
+
"grad_norm": 6.46875,
|
| 408 |
+
"learning_rate": 7.425876010781672e-06,
|
| 409 |
+
"loss": 0.28182178497314453,
|
| 410 |
+
"step": 2800
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 1.280323450134771,
|
| 414 |
+
"grad_norm": 7.15625,
|
| 415 |
+
"learning_rate": 7.20125786163522e-06,
|
| 416 |
+
"loss": 0.2626816368103027,
|
| 417 |
+
"step": 2850
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 1.302785265049416,
|
| 421 |
+
"grad_norm": 10.4375,
|
| 422 |
+
"learning_rate": 6.97663971248877e-06,
|
| 423 |
+
"loss": 0.36117866516113284,
|
| 424 |
+
"step": 2900
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 1.3252470799640612,
|
| 428 |
+
"grad_norm": 10.1875,
|
| 429 |
+
"learning_rate": 6.752021563342318e-06,
|
| 430 |
+
"loss": 0.2978287506103516,
|
| 431 |
+
"step": 2950
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 1.3477088948787062,
|
| 435 |
+
"grad_norm": 7.375,
|
| 436 |
+
"learning_rate": 6.527403414195867e-06,
|
| 437 |
+
"loss": 0.3120163917541504,
|
| 438 |
+
"step": 3000
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 1.3701707097933513,
|
| 442 |
+
"grad_norm": 3.421875,
|
| 443 |
+
"learning_rate": 6.302785265049417e-06,
|
| 444 |
+
"loss": 0.25403877258300783,
|
| 445 |
+
"step": 3050
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 1.3926325247079965,
|
| 449 |
+
"grad_norm": 11.8125,
|
| 450 |
+
"learning_rate": 6.078167115902965e-06,
|
| 451 |
+
"loss": 0.33552371978759765,
|
| 452 |
+
"step": 3100
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 1.4150943396226414,
|
| 456 |
+
"grad_norm": 10.8125,
|
| 457 |
+
"learning_rate": 5.853548966756515e-06,
|
| 458 |
+
"loss": 0.3004562759399414,
|
| 459 |
+
"step": 3150
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 1.4375561545372866,
|
| 463 |
+
"grad_norm": 16.0,
|
| 464 |
+
"learning_rate": 5.6289308176100635e-06,
|
| 465 |
+
"loss": 0.32779640197753906,
|
| 466 |
+
"step": 3200
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 1.4600179694519317,
|
| 470 |
+
"grad_norm": 15.0,
|
| 471 |
+
"learning_rate": 5.404312668463612e-06,
|
| 472 |
+
"loss": 0.2941103172302246,
|
| 473 |
+
"step": 3250
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 1.482479784366577,
|
| 477 |
+
"grad_norm": 10.8125,
|
| 478 |
+
"learning_rate": 5.179694519317161e-06,
|
| 479 |
+
"loss": 0.2714033889770508,
|
| 480 |
+
"step": 3300
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 1.504941599281222,
|
| 484 |
+
"grad_norm": 9.9375,
|
| 485 |
+
"learning_rate": 4.95507637017071e-06,
|
| 486 |
+
"loss": 0.3396564865112305,
|
| 487 |
+
"step": 3350
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 1.527403414195867,
|
| 491 |
+
"grad_norm": 7.8125,
|
| 492 |
+
"learning_rate": 4.730458221024259e-06,
|
| 493 |
+
"loss": 0.3063199996948242,
|
| 494 |
+
"step": 3400
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 1.5498652291105122,
|
| 498 |
+
"grad_norm": 9.25,
|
| 499 |
+
"learning_rate": 4.5058400718778085e-06,
|
| 500 |
+
"loss": 0.2858513641357422,
|
| 501 |
+
"step": 3450
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 1.5723270440251573,
|
| 505 |
+
"grad_norm": 11.0625,
|
| 506 |
+
"learning_rate": 4.281221922731357e-06,
|
| 507 |
+
"loss": 0.2822472381591797,
|
| 508 |
+
"step": 3500
|
| 509 |
+
},
|
| 510 |
+
{
|
| 511 |
+
"epoch": 1.5947888589398023,
|
| 512 |
+
"grad_norm": 8.0625,
|
| 513 |
+
"learning_rate": 4.056603773584906e-06,
|
| 514 |
+
"loss": 0.30435176849365236,
|
| 515 |
+
"step": 3550
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 1.6172506738544474,
|
| 519 |
+
"grad_norm": 12.125,
|
| 520 |
+
"learning_rate": 3.831985624438455e-06,
|
| 521 |
+
"loss": 0.31130361557006836,
|
| 522 |
+
"step": 3600
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 1.6397124887690926,
|
| 526 |
+
"grad_norm": 10.8125,
|
| 527 |
+
"learning_rate": 3.607367475292004e-06,
|
| 528 |
+
"loss": 0.34414665222167967,
|
| 529 |
+
"step": 3650
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 1.6621743036837375,
|
| 533 |
+
"grad_norm": 12.5625,
|
| 534 |
+
"learning_rate": 3.382749326145553e-06,
|
| 535 |
+
"loss": 0.297170524597168,
|
| 536 |
+
"step": 3700
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 1.684636118598383,
|
| 540 |
+
"grad_norm": 8.25,
|
| 541 |
+
"learning_rate": 3.158131176999102e-06,
|
| 542 |
+
"loss": 0.33300048828125,
|
| 543 |
+
"step": 3750
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 1.7070979335130279,
|
| 547 |
+
"grad_norm": 12.0,
|
| 548 |
+
"learning_rate": 2.9335130278526508e-06,
|
| 549 |
+
"loss": 0.3089234161376953,
|
| 550 |
+
"step": 3800
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"epoch": 1.7295597484276728,
|
| 554 |
+
"grad_norm": 11.6875,
|
| 555 |
+
"learning_rate": 2.7088948787061996e-06,
|
| 556 |
+
"loss": 0.320355224609375,
|
| 557 |
+
"step": 3850
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"epoch": 1.7520215633423182,
|
| 561 |
+
"grad_norm": 10.75,
|
| 562 |
+
"learning_rate": 2.4842767295597484e-06,
|
| 563 |
+
"loss": 0.32207695007324216,
|
| 564 |
+
"step": 3900
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 1.7744833782569631,
|
| 568 |
+
"grad_norm": 11.75,
|
| 569 |
+
"learning_rate": 2.2596585804132977e-06,
|
| 570 |
+
"loss": 0.33140968322753905,
|
| 571 |
+
"step": 3950
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 1.7969451931716083,
|
| 575 |
+
"grad_norm": 14.25,
|
| 576 |
+
"learning_rate": 2.0350404312668465e-06,
|
| 577 |
+
"loss": 0.2749526023864746,
|
| 578 |
+
"step": 4000
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 1.8194070080862534,
|
| 582 |
+
"grad_norm": 7.28125,
|
| 583 |
+
"learning_rate": 1.8104222821203954e-06,
|
| 584 |
+
"loss": 0.326712532043457,
|
| 585 |
+
"step": 4050
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 1.8418688230008984,
|
| 589 |
+
"grad_norm": 17.25,
|
| 590 |
+
"learning_rate": 1.5858041329739444e-06,
|
| 591 |
+
"loss": 0.3068280410766602,
|
| 592 |
+
"step": 4100
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 1.8643306379155435,
|
| 596 |
+
"grad_norm": 11.5625,
|
| 597 |
+
"learning_rate": 1.3611859838274934e-06,
|
| 598 |
+
"loss": 0.2806360626220703,
|
| 599 |
+
"step": 4150
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 1.8867924528301887,
|
| 603 |
+
"grad_norm": 18.75,
|
| 604 |
+
"learning_rate": 1.1365678346810423e-06,
|
| 605 |
+
"loss": 0.3010626411437988,
|
| 606 |
+
"step": 4200
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"epoch": 1.9092542677448336,
|
| 610 |
+
"grad_norm": 12.1875,
|
| 611 |
+
"learning_rate": 9.119496855345912e-07,
|
| 612 |
+
"loss": 0.30144317626953127,
|
| 613 |
+
"step": 4250
|
| 614 |
+
},
|
| 615 |
+
{
|
| 616 |
+
"epoch": 1.931716082659479,
|
| 617 |
+
"grad_norm": 7.21875,
|
| 618 |
+
"learning_rate": 6.873315363881403e-07,
|
| 619 |
+
"loss": 0.34434162139892577,
|
| 620 |
+
"step": 4300
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 1.954177897574124,
|
| 624 |
+
"grad_norm": 10.0625,
|
| 625 |
+
"learning_rate": 4.627133872416892e-07,
|
| 626 |
+
"loss": 0.32250152587890624,
|
| 627 |
+
"step": 4350
|
| 628 |
+
},
|
| 629 |
+
{
|
| 630 |
+
"epoch": 1.9766397124887691,
|
| 631 |
+
"grad_norm": 11.125,
|
| 632 |
+
"learning_rate": 2.3809523809523811e-07,
|
| 633 |
+
"loss": 0.276928596496582,
|
| 634 |
+
"step": 4400
|
| 635 |
+
},
|
| 636 |
+
{
|
| 637 |
+
"epoch": 1.9991015274034143,
|
| 638 |
+
"grad_norm": 9.9375,
|
| 639 |
+
"learning_rate": 1.3477088948787064e-08,
|
| 640 |
+
"loss": 0.29461381912231444,
|
| 641 |
+
"step": 4450
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 2.0,
|
| 645 |
+
"eval_loss": 0.3810727298259735,
|
| 646 |
+
"eval_runtime": 1.5694,
|
| 647 |
+
"eval_samples_per_second": 215.374,
|
| 648 |
+
"eval_steps_per_second": 7.009,
|
| 649 |
+
"step": 4452
|
| 650 |
+
}
|
| 651 |
+
],
|
| 652 |
+
"logging_steps": 50,
|
| 653 |
+
"max_steps": 4452,
|
| 654 |
+
"num_input_tokens_seen": 0,
|
| 655 |
+
"num_train_epochs": 2,
|
| 656 |
+
"save_steps": 500,
|
| 657 |
+
"stateful_callbacks": {
|
| 658 |
+
"TrainerControl": {
|
| 659 |
+
"args": {
|
| 660 |
+
"should_epoch_stop": false,
|
| 661 |
+
"should_evaluate": false,
|
| 662 |
+
"should_log": false,
|
| 663 |
+
"should_save": true,
|
| 664 |
+
"should_training_stop": true
|
| 665 |
+
},
|
| 666 |
+
"attributes": {}
|
| 667 |
+
}
|
| 668 |
+
},
|
| 669 |
+
"total_flos": 0.0,
|
| 670 |
+
"train_batch_size": 16,
|
| 671 |
+
"trial_name": null,
|
| 672 |
+
"trial_params": null
|
| 673 |
+
}
|
nboost-pt-biobert-base-msmarco-E2-S1-Mpairwise-FullDataTrue/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ba69ee7e1febfef9055f6a1a9759374537359d320d7ce3f68cb864303f36bae1
|
| 3 |
+
size 5329
|