Commit
·
efc3f4c
1
Parent(s):
dbed530
new model commit
Browse files- checkpoint-3500/config.json +74 -0
- checkpoint-3500/merges.txt +0 -0
- checkpoint-3500/optimizer.pt +3 -0
- checkpoint-3500/pytorch_model.bin +3 -0
- checkpoint-3500/rng_state.pth +3 -0
- checkpoint-3500/scheduler.pt +3 -0
- checkpoint-3500/special_tokens_map.json +15 -0
- checkpoint-3500/tokenizer.json +0 -0
- checkpoint-3500/tokenizer_config.json +16 -0
- checkpoint-3500/trainer_state.json +436 -0
- checkpoint-3500/training_args.bin +3 -0
- checkpoint-3500/vocab.json +0 -0
checkpoint-3500/config.json
ADDED
|
@@ -0,0 +1,74 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "mse30/bart-base-finetuned-pubmed",
|
| 3 |
+
"activation_dropout": 0.1,
|
| 4 |
+
"activation_function": "gelu",
|
| 5 |
+
"add_bias_logits": false,
|
| 6 |
+
"add_final_layer_norm": false,
|
| 7 |
+
"architectures": [
|
| 8 |
+
"BartForConditionalGeneration"
|
| 9 |
+
],
|
| 10 |
+
"attention_dropout": 0.1,
|
| 11 |
+
"bos_token_id": 0,
|
| 12 |
+
"classif_dropout": 0.1,
|
| 13 |
+
"classifier_dropout": 0.0,
|
| 14 |
+
"d_model": 768,
|
| 15 |
+
"decoder_attention_heads": 12,
|
| 16 |
+
"decoder_ffn_dim": 3072,
|
| 17 |
+
"decoder_layerdrop": 0.0,
|
| 18 |
+
"decoder_layers": 6,
|
| 19 |
+
"decoder_start_token_id": 2,
|
| 20 |
+
"dropout": 0.1,
|
| 21 |
+
"early_stopping": true,
|
| 22 |
+
"encoder_attention_heads": 12,
|
| 23 |
+
"encoder_ffn_dim": 3072,
|
| 24 |
+
"encoder_layerdrop": 0.0,
|
| 25 |
+
"encoder_layers": 6,
|
| 26 |
+
"eos_token_id": 2,
|
| 27 |
+
"forced_eos_token_id": 2,
|
| 28 |
+
"gradient_checkpointing": false,
|
| 29 |
+
"id2label": {
|
| 30 |
+
"0": "LABEL_0",
|
| 31 |
+
"1": "LABEL_1",
|
| 32 |
+
"2": "LABEL_2"
|
| 33 |
+
},
|
| 34 |
+
"init_std": 0.02,
|
| 35 |
+
"is_encoder_decoder": true,
|
| 36 |
+
"label2id": {
|
| 37 |
+
"LABEL_0": 0,
|
| 38 |
+
"LABEL_1": 1,
|
| 39 |
+
"LABEL_2": 2
|
| 40 |
+
},
|
| 41 |
+
"max_position_embeddings": 1024,
|
| 42 |
+
"model_type": "bart",
|
| 43 |
+
"no_repeat_ngram_size": 3,
|
| 44 |
+
"normalize_before": false,
|
| 45 |
+
"normalize_embedding": true,
|
| 46 |
+
"num_beams": 4,
|
| 47 |
+
"num_hidden_layers": 6,
|
| 48 |
+
"pad_token_id": 1,
|
| 49 |
+
"scale_embedding": false,
|
| 50 |
+
"task_specific_params": {
|
| 51 |
+
"summarization": {
|
| 52 |
+
"length_penalty": 1.0,
|
| 53 |
+
"max_length": 128,
|
| 54 |
+
"min_length": 12,
|
| 55 |
+
"num_beams": 4
|
| 56 |
+
},
|
| 57 |
+
"summarization_cnn": {
|
| 58 |
+
"length_penalty": 2.0,
|
| 59 |
+
"max_length": 142,
|
| 60 |
+
"min_length": 56,
|
| 61 |
+
"num_beams": 4
|
| 62 |
+
},
|
| 63 |
+
"summarization_xsum": {
|
| 64 |
+
"length_penalty": 1.0,
|
| 65 |
+
"max_length": 62,
|
| 66 |
+
"min_length": 11,
|
| 67 |
+
"num_beams": 6
|
| 68 |
+
}
|
| 69 |
+
},
|
| 70 |
+
"torch_dtype": "float32",
|
| 71 |
+
"transformers_version": "4.20.1",
|
| 72 |
+
"use_cache": true,
|
| 73 |
+
"vocab_size": 50265
|
| 74 |
+
}
|
checkpoint-3500/merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-3500/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:742248ffb56c93cd0311d0eee99436f2f1ad2f93c4e639cb6995b2a5c91ed5f3
|
| 3 |
+
size 1115513717
|
checkpoint-3500/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8b23a049afadf431ec6a2cf7b3cb42946d388be43fc5345e399143867f7491f0
|
| 3 |
+
size 557969145
|
checkpoint-3500/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e97c96fe8a8a62f4f8098428b6ec491ebe59e85cf041e90200cd4640d21733e3
|
| 3 |
+
size 14439
|
checkpoint-3500/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c78988eae64dcd9bb9caf2235a0b9cb59caaceb857c2958454f8b519c097dd89
|
| 3 |
+
size 623
|
checkpoint-3500/special_tokens_map.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": "<s>",
|
| 3 |
+
"cls_token": "<s>",
|
| 4 |
+
"eos_token": "</s>",
|
| 5 |
+
"mask_token": {
|
| 6 |
+
"content": "<mask>",
|
| 7 |
+
"lstrip": true,
|
| 8 |
+
"normalized": false,
|
| 9 |
+
"rstrip": false,
|
| 10 |
+
"single_word": false
|
| 11 |
+
},
|
| 12 |
+
"pad_token": "<pad>",
|
| 13 |
+
"sep_token": "</s>",
|
| 14 |
+
"unk_token": "<unk>"
|
| 15 |
+
}
|
checkpoint-3500/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-3500/tokenizer_config.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"bos_token": "<s>",
|
| 4 |
+
"cls_token": "<s>",
|
| 5 |
+
"eos_token": "</s>",
|
| 6 |
+
"errors": "replace",
|
| 7 |
+
"mask_token": "<mask>",
|
| 8 |
+
"model_max_length": 1024,
|
| 9 |
+
"name_or_path": "mse30/bart-base-finetuned-pubmed",
|
| 10 |
+
"pad_token": "<pad>",
|
| 11 |
+
"sep_token": "</s>",
|
| 12 |
+
"special_tokens_map_file": null,
|
| 13 |
+
"tokenizer_class": "BartTokenizer",
|
| 14 |
+
"trim_offsets": true,
|
| 15 |
+
"unk_token": "<unk>"
|
| 16 |
+
}
|
checkpoint-3500/trainer_state.json
ADDED
|
@@ -0,0 +1,436 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 97.22222222222223,
|
| 5 |
+
"global_step": 3500,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 1.39,
|
| 12 |
+
"learning_rate": 3e-06,
|
| 13 |
+
"loss": 7.0999,
|
| 14 |
+
"step": 50
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 2.78,
|
| 18 |
+
"learning_rate": 6e-06,
|
| 19 |
+
"loss": 3.9223,
|
| 20 |
+
"step": 100
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 4.17,
|
| 24 |
+
"learning_rate": 9e-06,
|
| 25 |
+
"loss": 2.5274,
|
| 26 |
+
"step": 150
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 5.56,
|
| 30 |
+
"learning_rate": 1.2e-05,
|
| 31 |
+
"loss": 2.1012,
|
| 32 |
+
"step": 200
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 6.94,
|
| 36 |
+
"learning_rate": 1.5e-05,
|
| 37 |
+
"loss": 1.9323,
|
| 38 |
+
"step": 250
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 8.33,
|
| 42 |
+
"learning_rate": 1.8e-05,
|
| 43 |
+
"loss": 1.7811,
|
| 44 |
+
"step": 300
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 9.72,
|
| 48 |
+
"learning_rate": 2.1e-05,
|
| 49 |
+
"loss": 1.7308,
|
| 50 |
+
"step": 350
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 11.11,
|
| 54 |
+
"learning_rate": 2.4e-05,
|
| 55 |
+
"loss": 1.6515,
|
| 56 |
+
"step": 400
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 12.5,
|
| 60 |
+
"learning_rate": 2.7000000000000002e-05,
|
| 61 |
+
"loss": 1.6299,
|
| 62 |
+
"step": 450
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 13.89,
|
| 66 |
+
"learning_rate": 3e-05,
|
| 67 |
+
"loss": 1.6019,
|
| 68 |
+
"step": 500
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 15.28,
|
| 72 |
+
"learning_rate": 2.9516129032258067e-05,
|
| 73 |
+
"loss": 1.5727,
|
| 74 |
+
"step": 550
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 16.67,
|
| 78 |
+
"learning_rate": 2.903225806451613e-05,
|
| 79 |
+
"loss": 1.5442,
|
| 80 |
+
"step": 600
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 18.06,
|
| 84 |
+
"learning_rate": 2.8548387096774196e-05,
|
| 85 |
+
"loss": 1.5382,
|
| 86 |
+
"step": 650
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 19.44,
|
| 90 |
+
"learning_rate": 2.806451612903226e-05,
|
| 91 |
+
"loss": 1.5224,
|
| 92 |
+
"step": 700
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 20.83,
|
| 96 |
+
"learning_rate": 2.758064516129032e-05,
|
| 97 |
+
"loss": 1.5323,
|
| 98 |
+
"step": 750
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 22.22,
|
| 102 |
+
"learning_rate": 2.7096774193548387e-05,
|
| 103 |
+
"loss": 1.5036,
|
| 104 |
+
"step": 800
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 23.61,
|
| 108 |
+
"learning_rate": 2.6612903225806453e-05,
|
| 109 |
+
"loss": 1.4981,
|
| 110 |
+
"step": 850
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 25.0,
|
| 114 |
+
"learning_rate": 2.6129032258064516e-05,
|
| 115 |
+
"loss": 1.5055,
|
| 116 |
+
"step": 900
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 26.39,
|
| 120 |
+
"learning_rate": 2.5645161290322582e-05,
|
| 121 |
+
"loss": 1.4892,
|
| 122 |
+
"step": 950
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 27.78,
|
| 126 |
+
"learning_rate": 2.5161290322580648e-05,
|
| 127 |
+
"loss": 1.4892,
|
| 128 |
+
"step": 1000
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 29.17,
|
| 132 |
+
"learning_rate": 2.467741935483871e-05,
|
| 133 |
+
"loss": 1.4802,
|
| 134 |
+
"step": 1050
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 30.56,
|
| 138 |
+
"learning_rate": 2.4193548387096773e-05,
|
| 139 |
+
"loss": 1.4804,
|
| 140 |
+
"step": 1100
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 31.94,
|
| 144 |
+
"learning_rate": 2.370967741935484e-05,
|
| 145 |
+
"loss": 1.4692,
|
| 146 |
+
"step": 1150
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 33.33,
|
| 150 |
+
"learning_rate": 2.3225806451612902e-05,
|
| 151 |
+
"loss": 1.478,
|
| 152 |
+
"step": 1200
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 34.72,
|
| 156 |
+
"learning_rate": 2.274193548387097e-05,
|
| 157 |
+
"loss": 1.4692,
|
| 158 |
+
"step": 1250
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 36.11,
|
| 162 |
+
"learning_rate": 2.2258064516129034e-05,
|
| 163 |
+
"loss": 1.465,
|
| 164 |
+
"step": 1300
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 37.5,
|
| 168 |
+
"learning_rate": 2.1774193548387097e-05,
|
| 169 |
+
"loss": 1.4653,
|
| 170 |
+
"step": 1350
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 38.89,
|
| 174 |
+
"learning_rate": 2.1290322580645163e-05,
|
| 175 |
+
"loss": 1.4636,
|
| 176 |
+
"step": 1400
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 40.28,
|
| 180 |
+
"learning_rate": 2.080645161290323e-05,
|
| 181 |
+
"loss": 1.467,
|
| 182 |
+
"step": 1450
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 41.67,
|
| 186 |
+
"learning_rate": 2.032258064516129e-05,
|
| 187 |
+
"loss": 1.4605,
|
| 188 |
+
"step": 1500
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 43.06,
|
| 192 |
+
"learning_rate": 1.9838709677419355e-05,
|
| 193 |
+
"loss": 1.4617,
|
| 194 |
+
"step": 1550
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 44.44,
|
| 198 |
+
"learning_rate": 1.935483870967742e-05,
|
| 199 |
+
"loss": 1.4632,
|
| 200 |
+
"step": 1600
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 45.83,
|
| 204 |
+
"learning_rate": 1.8870967741935484e-05,
|
| 205 |
+
"loss": 1.4641,
|
| 206 |
+
"step": 1650
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 47.22,
|
| 210 |
+
"learning_rate": 1.838709677419355e-05,
|
| 211 |
+
"loss": 1.4528,
|
| 212 |
+
"step": 1700
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 48.61,
|
| 216 |
+
"learning_rate": 1.7903225806451616e-05,
|
| 217 |
+
"loss": 1.4566,
|
| 218 |
+
"step": 1750
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 50.0,
|
| 222 |
+
"learning_rate": 1.741935483870968e-05,
|
| 223 |
+
"loss": 1.4522,
|
| 224 |
+
"step": 1800
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 51.39,
|
| 228 |
+
"learning_rate": 1.6935483870967744e-05,
|
| 229 |
+
"loss": 1.4598,
|
| 230 |
+
"step": 1850
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 52.78,
|
| 234 |
+
"learning_rate": 1.6451612903225807e-05,
|
| 235 |
+
"loss": 1.454,
|
| 236 |
+
"step": 1900
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 54.17,
|
| 240 |
+
"learning_rate": 1.596774193548387e-05,
|
| 241 |
+
"loss": 1.4568,
|
| 242 |
+
"step": 1950
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 55.56,
|
| 246 |
+
"learning_rate": 1.5483870967741936e-05,
|
| 247 |
+
"loss": 1.4528,
|
| 248 |
+
"step": 2000
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 56.94,
|
| 252 |
+
"learning_rate": 1.5e-05,
|
| 253 |
+
"loss": 1.4471,
|
| 254 |
+
"step": 2050
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 58.33,
|
| 258 |
+
"learning_rate": 1.4516129032258065e-05,
|
| 259 |
+
"loss": 1.4471,
|
| 260 |
+
"step": 2100
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 59.72,
|
| 264 |
+
"learning_rate": 1.403225806451613e-05,
|
| 265 |
+
"loss": 1.4554,
|
| 266 |
+
"step": 2150
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 61.11,
|
| 270 |
+
"learning_rate": 1.3548387096774194e-05,
|
| 271 |
+
"loss": 1.4435,
|
| 272 |
+
"step": 2200
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 62.5,
|
| 276 |
+
"learning_rate": 1.3064516129032258e-05,
|
| 277 |
+
"loss": 1.4456,
|
| 278 |
+
"step": 2250
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 63.89,
|
| 282 |
+
"learning_rate": 1.2580645161290324e-05,
|
| 283 |
+
"loss": 1.4452,
|
| 284 |
+
"step": 2300
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 65.28,
|
| 288 |
+
"learning_rate": 1.2096774193548387e-05,
|
| 289 |
+
"loss": 1.4496,
|
| 290 |
+
"step": 2350
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 66.67,
|
| 294 |
+
"learning_rate": 1.1612903225806451e-05,
|
| 295 |
+
"loss": 1.4431,
|
| 296 |
+
"step": 2400
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 68.06,
|
| 300 |
+
"learning_rate": 1.1129032258064517e-05,
|
| 301 |
+
"loss": 1.4433,
|
| 302 |
+
"step": 2450
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 69.44,
|
| 306 |
+
"learning_rate": 1.0645161290322582e-05,
|
| 307 |
+
"loss": 1.4437,
|
| 308 |
+
"step": 2500
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 70.83,
|
| 312 |
+
"learning_rate": 1.0161290322580644e-05,
|
| 313 |
+
"loss": 1.4414,
|
| 314 |
+
"step": 2550
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 72.22,
|
| 318 |
+
"learning_rate": 9.67741935483871e-06,
|
| 319 |
+
"loss": 1.4413,
|
| 320 |
+
"step": 2600
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 73.61,
|
| 324 |
+
"learning_rate": 9.193548387096775e-06,
|
| 325 |
+
"loss": 1.4414,
|
| 326 |
+
"step": 2650
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 75.0,
|
| 330 |
+
"learning_rate": 8.70967741935484e-06,
|
| 331 |
+
"loss": 1.4396,
|
| 332 |
+
"step": 2700
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 76.39,
|
| 336 |
+
"learning_rate": 8.225806451612904e-06,
|
| 337 |
+
"loss": 1.4415,
|
| 338 |
+
"step": 2750
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 77.78,
|
| 342 |
+
"learning_rate": 7.741935483870968e-06,
|
| 343 |
+
"loss": 1.443,
|
| 344 |
+
"step": 2800
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 79.17,
|
| 348 |
+
"learning_rate": 7.258064516129032e-06,
|
| 349 |
+
"loss": 1.4432,
|
| 350 |
+
"step": 2850
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 80.56,
|
| 354 |
+
"learning_rate": 6.774193548387097e-06,
|
| 355 |
+
"loss": 1.4389,
|
| 356 |
+
"step": 2900
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 81.94,
|
| 360 |
+
"learning_rate": 6.290322580645162e-06,
|
| 361 |
+
"loss": 1.4383,
|
| 362 |
+
"step": 2950
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 83.33,
|
| 366 |
+
"learning_rate": 5.8064516129032256e-06,
|
| 367 |
+
"loss": 1.4387,
|
| 368 |
+
"step": 3000
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 84.72,
|
| 372 |
+
"learning_rate": 5.322580645161291e-06,
|
| 373 |
+
"loss": 1.4389,
|
| 374 |
+
"step": 3050
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 86.11,
|
| 378 |
+
"learning_rate": 4.838709677419355e-06,
|
| 379 |
+
"loss": 1.4374,
|
| 380 |
+
"step": 3100
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 87.5,
|
| 384 |
+
"learning_rate": 4.35483870967742e-06,
|
| 385 |
+
"loss": 1.4368,
|
| 386 |
+
"step": 3150
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 88.89,
|
| 390 |
+
"learning_rate": 3.870967741935484e-06,
|
| 391 |
+
"loss": 1.442,
|
| 392 |
+
"step": 3200
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 90.28,
|
| 396 |
+
"learning_rate": 3.3870967741935484e-06,
|
| 397 |
+
"loss": 1.4385,
|
| 398 |
+
"step": 3250
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 91.67,
|
| 402 |
+
"learning_rate": 2.9032258064516128e-06,
|
| 403 |
+
"loss": 1.4395,
|
| 404 |
+
"step": 3300
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 93.06,
|
| 408 |
+
"learning_rate": 2.4193548387096776e-06,
|
| 409 |
+
"loss": 1.4396,
|
| 410 |
+
"step": 3350
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 94.44,
|
| 414 |
+
"learning_rate": 1.935483870967742e-06,
|
| 415 |
+
"loss": 1.4372,
|
| 416 |
+
"step": 3400
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 95.83,
|
| 420 |
+
"learning_rate": 1.4516129032258064e-06,
|
| 421 |
+
"loss": 1.4359,
|
| 422 |
+
"step": 3450
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 97.22,
|
| 426 |
+
"learning_rate": 9.67741935483871e-07,
|
| 427 |
+
"loss": 1.4376,
|
| 428 |
+
"step": 3500
|
| 429 |
+
}
|
| 430 |
+
],
|
| 431 |
+
"max_steps": 3600,
|
| 432 |
+
"num_train_epochs": 100,
|
| 433 |
+
"total_flos": 8418021542461440.0,
|
| 434 |
+
"trial_name": null,
|
| 435 |
+
"trial_params": null
|
| 436 |
+
}
|
checkpoint-3500/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b4be4af3d3f53d27480b180dfb2b1d5ebb63a895d4a2c8a5ca7f3b04d4de1d78
|
| 3 |
+
size 3375
|
checkpoint-3500/vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|