Earlan Josh Sabillano
commited on
Upload folder using huggingface_hub
Browse files- pet-collaboration-model-facebook-bart-large-mnli_bs4_gas4_lr5e-05_ep3/checkpoint-936/config.json +51 -0
- pet-collaboration-model-facebook-bart-large-mnli_bs4_gas4_lr5e-05_ep3/checkpoint-936/generation_config.json +9 -0
- pet-collaboration-model-facebook-bart-large-mnli_bs4_gas4_lr5e-05_ep3/checkpoint-936/model.safetensors +3 -0
- pet-collaboration-model-facebook-bart-large-mnli_bs4_gas4_lr5e-05_ep3/checkpoint-936/optimizer.pt +3 -0
- pet-collaboration-model-facebook-bart-large-mnli_bs4_gas4_lr5e-05_ep3/checkpoint-936/rng_state.pth +3 -0
- pet-collaboration-model-facebook-bart-large-mnli_bs4_gas4_lr5e-05_ep3/checkpoint-936/scheduler.pt +3 -0
- pet-collaboration-model-facebook-bart-large-mnli_bs4_gas4_lr5e-05_ep3/checkpoint-936/trainer_state.json +708 -0
- pet-collaboration-model-facebook-bart-large-mnli_bs4_gas4_lr5e-05_ep3/checkpoint-936/training_args.bin +3 -0
pet-collaboration-model-facebook-bart-large-mnli_bs4_gas4_lr5e-05_ep3/checkpoint-936/config.json
ADDED
|
@@ -0,0 +1,51 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "facebook/bart-large-mnli",
|
| 3 |
+
"_num_labels": 3,
|
| 4 |
+
"activation_dropout": 0.0,
|
| 5 |
+
"activation_function": "gelu",
|
| 6 |
+
"add_final_layer_norm": false,
|
| 7 |
+
"architectures": [
|
| 8 |
+
"BartForConditionalGeneration"
|
| 9 |
+
],
|
| 10 |
+
"attention_dropout": 0.0,
|
| 11 |
+
"bos_token_id": 0,
|
| 12 |
+
"classif_dropout": 0.0,
|
| 13 |
+
"classifier_dropout": 0.0,
|
| 14 |
+
"d_model": 1024,
|
| 15 |
+
"decoder_attention_heads": 16,
|
| 16 |
+
"decoder_ffn_dim": 4096,
|
| 17 |
+
"decoder_layerdrop": 0.0,
|
| 18 |
+
"decoder_layers": 12,
|
| 19 |
+
"decoder_start_token_id": 2,
|
| 20 |
+
"dropout": 0.1,
|
| 21 |
+
"encoder_attention_heads": 16,
|
| 22 |
+
"encoder_ffn_dim": 4096,
|
| 23 |
+
"encoder_layerdrop": 0.0,
|
| 24 |
+
"encoder_layers": 12,
|
| 25 |
+
"eos_token_id": 2,
|
| 26 |
+
"forced_eos_token_id": 2,
|
| 27 |
+
"gradient_checkpointing": false,
|
| 28 |
+
"id2label": {
|
| 29 |
+
"0": "contradiction",
|
| 30 |
+
"1": "neutral",
|
| 31 |
+
"2": "entailment"
|
| 32 |
+
},
|
| 33 |
+
"init_std": 0.02,
|
| 34 |
+
"is_encoder_decoder": true,
|
| 35 |
+
"label2id": {
|
| 36 |
+
"contradiction": 0,
|
| 37 |
+
"entailment": 2,
|
| 38 |
+
"neutral": 1
|
| 39 |
+
},
|
| 40 |
+
"max_position_embeddings": 1024,
|
| 41 |
+
"model_type": "bart",
|
| 42 |
+
"normalize_before": false,
|
| 43 |
+
"num_hidden_layers": 12,
|
| 44 |
+
"output_past": false,
|
| 45 |
+
"pad_token_id": 1,
|
| 46 |
+
"scale_embedding": false,
|
| 47 |
+
"torch_dtype": "float32",
|
| 48 |
+
"transformers_version": "4.49.0",
|
| 49 |
+
"use_cache": true,
|
| 50 |
+
"vocab_size": 50265
|
| 51 |
+
}
|
pet-collaboration-model-facebook-bart-large-mnli_bs4_gas4_lr5e-05_ep3/checkpoint-936/generation_config.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 0,
|
| 4 |
+
"decoder_start_token_id": 2,
|
| 5 |
+
"eos_token_id": 2,
|
| 6 |
+
"forced_eos_token_id": 2,
|
| 7 |
+
"pad_token_id": 1,
|
| 8 |
+
"transformers_version": "4.49.0"
|
| 9 |
+
}
|
pet-collaboration-model-facebook-bart-large-mnli_bs4_gas4_lr5e-05_ep3/checkpoint-936/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d2c8652725fc7c65df7fdfbbc508386233487c919d03be8aab9f22cc0507c704
|
| 3 |
+
size 1625426996
|
pet-collaboration-model-facebook-bart-large-mnli_bs4_gas4_lr5e-05_ep3/checkpoint-936/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b3f470f8602f86b0fa548f5baee6a69652ea27e9e3ee1c9ee5845fc96aee258c
|
| 3 |
+
size 3250759951
|
pet-collaboration-model-facebook-bart-large-mnli_bs4_gas4_lr5e-05_ep3/checkpoint-936/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:239cf6b7aaed07a9c18a05072d6057192da0d027dd0d202f1442aed732ef3b81
|
| 3 |
+
size 14244
|
pet-collaboration-model-facebook-bart-large-mnli_bs4_gas4_lr5e-05_ep3/checkpoint-936/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:68324c5f3555717a2b93f0f24586599046985b46500cc1930b51d4c09662924a
|
| 3 |
+
size 1064
|
pet-collaboration-model-facebook-bart-large-mnli_bs4_gas4_lr5e-05_ep3/checkpoint-936/trainer_state.json
ADDED
|
@@ -0,0 +1,708 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 936,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.03205128205128205,
|
| 13 |
+
"grad_norm": 26.609880447387695,
|
| 14 |
+
"learning_rate": 4.946581196581197e-05,
|
| 15 |
+
"loss": 2.9192,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.0641025641025641,
|
| 20 |
+
"grad_norm": 37.73039245605469,
|
| 21 |
+
"learning_rate": 4.893162393162393e-05,
|
| 22 |
+
"loss": 0.5855,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.09615384615384616,
|
| 27 |
+
"grad_norm": 5.29367733001709,
|
| 28 |
+
"learning_rate": 4.83974358974359e-05,
|
| 29 |
+
"loss": 0.7988,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.1282051282051282,
|
| 34 |
+
"grad_norm": 10.990591049194336,
|
| 35 |
+
"learning_rate": 4.786324786324787e-05,
|
| 36 |
+
"loss": 0.3698,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.16025641025641027,
|
| 41 |
+
"grad_norm": 26.886842727661133,
|
| 42 |
+
"learning_rate": 4.732905982905983e-05,
|
| 43 |
+
"loss": 0.3782,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.19230769230769232,
|
| 48 |
+
"grad_norm": 14.569891929626465,
|
| 49 |
+
"learning_rate": 4.67948717948718e-05,
|
| 50 |
+
"loss": 0.5114,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.22435897435897437,
|
| 55 |
+
"grad_norm": 2.1407554149627686,
|
| 56 |
+
"learning_rate": 4.626068376068376e-05,
|
| 57 |
+
"loss": 0.3389,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.2564102564102564,
|
| 62 |
+
"grad_norm": 2.868415594100952,
|
| 63 |
+
"learning_rate": 4.572649572649573e-05,
|
| 64 |
+
"loss": 0.2623,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.28846153846153844,
|
| 69 |
+
"grad_norm": 25.492149353027344,
|
| 70 |
+
"learning_rate": 4.519230769230769e-05,
|
| 71 |
+
"loss": 0.4462,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.32051282051282054,
|
| 76 |
+
"grad_norm": 7.5209808349609375,
|
| 77 |
+
"learning_rate": 4.465811965811966e-05,
|
| 78 |
+
"loss": 0.3118,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.3525641025641026,
|
| 83 |
+
"grad_norm": 17.091777801513672,
|
| 84 |
+
"learning_rate": 4.412393162393162e-05,
|
| 85 |
+
"loss": 0.442,
|
| 86 |
+
"step": 110
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.38461538461538464,
|
| 90 |
+
"grad_norm": 5.439161777496338,
|
| 91 |
+
"learning_rate": 4.358974358974359e-05,
|
| 92 |
+
"loss": 0.3757,
|
| 93 |
+
"step": 120
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.4166666666666667,
|
| 97 |
+
"grad_norm": 12.27194881439209,
|
| 98 |
+
"learning_rate": 4.305555555555556e-05,
|
| 99 |
+
"loss": 0.3616,
|
| 100 |
+
"step": 130
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.44871794871794873,
|
| 104 |
+
"grad_norm": 4.349429130554199,
|
| 105 |
+
"learning_rate": 4.252136752136752e-05,
|
| 106 |
+
"loss": 0.3499,
|
| 107 |
+
"step": 140
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.4807692307692308,
|
| 111 |
+
"grad_norm": 22.523239135742188,
|
| 112 |
+
"learning_rate": 4.198717948717949e-05,
|
| 113 |
+
"loss": 0.3438,
|
| 114 |
+
"step": 150
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.5128205128205128,
|
| 118 |
+
"grad_norm": 5.383206367492676,
|
| 119 |
+
"learning_rate": 4.145299145299146e-05,
|
| 120 |
+
"loss": 0.322,
|
| 121 |
+
"step": 160
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.5448717948717948,
|
| 125 |
+
"grad_norm": 3.995262384414673,
|
| 126 |
+
"learning_rate": 4.091880341880342e-05,
|
| 127 |
+
"loss": 0.4186,
|
| 128 |
+
"step": 170
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.5769230769230769,
|
| 132 |
+
"grad_norm": 2.75403094291687,
|
| 133 |
+
"learning_rate": 4.038461538461539e-05,
|
| 134 |
+
"loss": 0.334,
|
| 135 |
+
"step": 180
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.6089743589743589,
|
| 139 |
+
"grad_norm": 7.16841983795166,
|
| 140 |
+
"learning_rate": 3.985042735042736e-05,
|
| 141 |
+
"loss": 0.279,
|
| 142 |
+
"step": 190
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.6410256410256411,
|
| 146 |
+
"grad_norm": 9.7506103515625,
|
| 147 |
+
"learning_rate": 3.931623931623932e-05,
|
| 148 |
+
"loss": 0.3787,
|
| 149 |
+
"step": 200
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.6730769230769231,
|
| 153 |
+
"grad_norm": 7.993756294250488,
|
| 154 |
+
"learning_rate": 3.878205128205129e-05,
|
| 155 |
+
"loss": 0.2262,
|
| 156 |
+
"step": 210
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.7051282051282052,
|
| 160 |
+
"grad_norm": 6.005373954772949,
|
| 161 |
+
"learning_rate": 3.824786324786325e-05,
|
| 162 |
+
"loss": 0.4179,
|
| 163 |
+
"step": 220
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.7371794871794872,
|
| 167 |
+
"grad_norm": 14.525274276733398,
|
| 168 |
+
"learning_rate": 3.771367521367522e-05,
|
| 169 |
+
"loss": 0.3105,
|
| 170 |
+
"step": 230
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.7692307692307693,
|
| 174 |
+
"grad_norm": 5.759218692779541,
|
| 175 |
+
"learning_rate": 3.717948717948718e-05,
|
| 176 |
+
"loss": 0.3378,
|
| 177 |
+
"step": 240
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.8012820512820513,
|
| 181 |
+
"grad_norm": 11.445273399353027,
|
| 182 |
+
"learning_rate": 3.664529914529915e-05,
|
| 183 |
+
"loss": 0.4971,
|
| 184 |
+
"step": 250
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.8333333333333334,
|
| 188 |
+
"grad_norm": 4.408927917480469,
|
| 189 |
+
"learning_rate": 3.611111111111111e-05,
|
| 190 |
+
"loss": 0.3357,
|
| 191 |
+
"step": 260
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.8653846153846154,
|
| 195 |
+
"grad_norm": 3.518434524536133,
|
| 196 |
+
"learning_rate": 3.557692307692308e-05,
|
| 197 |
+
"loss": 0.273,
|
| 198 |
+
"step": 270
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.8974358974358975,
|
| 202 |
+
"grad_norm": 2.5430490970611572,
|
| 203 |
+
"learning_rate": 3.504273504273504e-05,
|
| 204 |
+
"loss": 0.3705,
|
| 205 |
+
"step": 280
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.9294871794871795,
|
| 209 |
+
"grad_norm": 11.406705856323242,
|
| 210 |
+
"learning_rate": 3.450854700854701e-05,
|
| 211 |
+
"loss": 0.2497,
|
| 212 |
+
"step": 290
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.9615384615384616,
|
| 216 |
+
"grad_norm": 15.890022277832031,
|
| 217 |
+
"learning_rate": 3.397435897435898e-05,
|
| 218 |
+
"loss": 0.3378,
|
| 219 |
+
"step": 300
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.9935897435897436,
|
| 223 |
+
"grad_norm": 3.5740270614624023,
|
| 224 |
+
"learning_rate": 3.344017094017094e-05,
|
| 225 |
+
"loss": 0.3236,
|
| 226 |
+
"step": 310
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 1.0,
|
| 230 |
+
"eval_loss": 0.2678295075893402,
|
| 231 |
+
"eval_runtime": 7.4566,
|
| 232 |
+
"eval_samples_per_second": 167.368,
|
| 233 |
+
"eval_steps_per_second": 41.842,
|
| 234 |
+
"step": 312
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 1.0256410256410255,
|
| 238 |
+
"grad_norm": 4.209905624389648,
|
| 239 |
+
"learning_rate": 3.290598290598291e-05,
|
| 240 |
+
"loss": 0.1984,
|
| 241 |
+
"step": 320
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 1.0576923076923077,
|
| 245 |
+
"grad_norm": 7.845182418823242,
|
| 246 |
+
"learning_rate": 3.2371794871794876e-05,
|
| 247 |
+
"loss": 0.2263,
|
| 248 |
+
"step": 330
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 1.0897435897435896,
|
| 252 |
+
"grad_norm": 6.887997150421143,
|
| 253 |
+
"learning_rate": 3.183760683760684e-05,
|
| 254 |
+
"loss": 0.2463,
|
| 255 |
+
"step": 340
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 1.1217948717948718,
|
| 259 |
+
"grad_norm": 15.53697681427002,
|
| 260 |
+
"learning_rate": 3.1303418803418806e-05,
|
| 261 |
+
"loss": 0.1008,
|
| 262 |
+
"step": 350
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 1.1538461538461537,
|
| 266 |
+
"grad_norm": 8.783138275146484,
|
| 267 |
+
"learning_rate": 3.0769230769230774e-05,
|
| 268 |
+
"loss": 0.2439,
|
| 269 |
+
"step": 360
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 1.185897435897436,
|
| 273 |
+
"grad_norm": 16.240093231201172,
|
| 274 |
+
"learning_rate": 3.0235042735042736e-05,
|
| 275 |
+
"loss": 0.2978,
|
| 276 |
+
"step": 370
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 1.217948717948718,
|
| 280 |
+
"grad_norm": 8.975017547607422,
|
| 281 |
+
"learning_rate": 2.97008547008547e-05,
|
| 282 |
+
"loss": 0.2484,
|
| 283 |
+
"step": 380
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 1.25,
|
| 287 |
+
"grad_norm": 2.210106611251831,
|
| 288 |
+
"learning_rate": 2.916666666666667e-05,
|
| 289 |
+
"loss": 0.1266,
|
| 290 |
+
"step": 390
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 1.282051282051282,
|
| 294 |
+
"grad_norm": 20.064651489257812,
|
| 295 |
+
"learning_rate": 2.863247863247863e-05,
|
| 296 |
+
"loss": 0.1563,
|
| 297 |
+
"step": 400
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 1.314102564102564,
|
| 301 |
+
"grad_norm": 1.90776526927948,
|
| 302 |
+
"learning_rate": 2.80982905982906e-05,
|
| 303 |
+
"loss": 0.2336,
|
| 304 |
+
"step": 410
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 1.3461538461538463,
|
| 308 |
+
"grad_norm": 6.855515480041504,
|
| 309 |
+
"learning_rate": 2.756410256410257e-05,
|
| 310 |
+
"loss": 0.3492,
|
| 311 |
+
"step": 420
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 1.3782051282051282,
|
| 315 |
+
"grad_norm": 14.22805118560791,
|
| 316 |
+
"learning_rate": 2.702991452991453e-05,
|
| 317 |
+
"loss": 0.2186,
|
| 318 |
+
"step": 430
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 1.4102564102564101,
|
| 322 |
+
"grad_norm": 3.9664502143859863,
|
| 323 |
+
"learning_rate": 2.64957264957265e-05,
|
| 324 |
+
"loss": 0.1362,
|
| 325 |
+
"step": 440
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 1.4423076923076923,
|
| 329 |
+
"grad_norm": 3.347365140914917,
|
| 330 |
+
"learning_rate": 2.5961538461538464e-05,
|
| 331 |
+
"loss": 0.2644,
|
| 332 |
+
"step": 450
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 1.4743589743589745,
|
| 336 |
+
"grad_norm": 10.483346939086914,
|
| 337 |
+
"learning_rate": 2.5427350427350426e-05,
|
| 338 |
+
"loss": 0.1658,
|
| 339 |
+
"step": 460
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 1.5064102564102564,
|
| 343 |
+
"grad_norm": 5.915305137634277,
|
| 344 |
+
"learning_rate": 2.4893162393162394e-05,
|
| 345 |
+
"loss": 0.2567,
|
| 346 |
+
"step": 470
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 1.5384615384615383,
|
| 350 |
+
"grad_norm": 1.7681890726089478,
|
| 351 |
+
"learning_rate": 2.435897435897436e-05,
|
| 352 |
+
"loss": 0.2054,
|
| 353 |
+
"step": 480
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 1.5705128205128205,
|
| 357 |
+
"grad_norm": 5.247648239135742,
|
| 358 |
+
"learning_rate": 2.3824786324786324e-05,
|
| 359 |
+
"loss": 0.1632,
|
| 360 |
+
"step": 490
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 1.6025641025641026,
|
| 364 |
+
"grad_norm": 16.242462158203125,
|
| 365 |
+
"learning_rate": 2.3290598290598293e-05,
|
| 366 |
+
"loss": 0.1974,
|
| 367 |
+
"step": 500
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 1.6346153846153846,
|
| 371 |
+
"grad_norm": 6.025589942932129,
|
| 372 |
+
"learning_rate": 2.2756410256410258e-05,
|
| 373 |
+
"loss": 0.3603,
|
| 374 |
+
"step": 510
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 1.6666666666666665,
|
| 378 |
+
"grad_norm": 12.795687675476074,
|
| 379 |
+
"learning_rate": 2.2222222222222223e-05,
|
| 380 |
+
"loss": 0.2313,
|
| 381 |
+
"step": 520
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 1.6987179487179487,
|
| 385 |
+
"grad_norm": 5.124413013458252,
|
| 386 |
+
"learning_rate": 2.168803418803419e-05,
|
| 387 |
+
"loss": 0.2344,
|
| 388 |
+
"step": 530
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 1.7307692307692308,
|
| 392 |
+
"grad_norm": 0.8152647614479065,
|
| 393 |
+
"learning_rate": 2.1153846153846154e-05,
|
| 394 |
+
"loss": 0.2517,
|
| 395 |
+
"step": 540
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 1.7628205128205128,
|
| 399 |
+
"grad_norm": 4.152542591094971,
|
| 400 |
+
"learning_rate": 2.061965811965812e-05,
|
| 401 |
+
"loss": 0.2224,
|
| 402 |
+
"step": 550
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 1.7948717948717947,
|
| 406 |
+
"grad_norm": 8.524950981140137,
|
| 407 |
+
"learning_rate": 2.0085470085470087e-05,
|
| 408 |
+
"loss": 0.2012,
|
| 409 |
+
"step": 560
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 1.8269230769230769,
|
| 413 |
+
"grad_norm": 2.0259006023406982,
|
| 414 |
+
"learning_rate": 1.9551282051282052e-05,
|
| 415 |
+
"loss": 0.1671,
|
| 416 |
+
"step": 570
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 1.858974358974359,
|
| 420 |
+
"grad_norm": 4.448575496673584,
|
| 421 |
+
"learning_rate": 1.9017094017094017e-05,
|
| 422 |
+
"loss": 0.1634,
|
| 423 |
+
"step": 580
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 1.891025641025641,
|
| 427 |
+
"grad_norm": 1.187767744064331,
|
| 428 |
+
"learning_rate": 1.8482905982905986e-05,
|
| 429 |
+
"loss": 0.1741,
|
| 430 |
+
"step": 590
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 1.9230769230769231,
|
| 434 |
+
"grad_norm": 0.37538495659828186,
|
| 435 |
+
"learning_rate": 1.794871794871795e-05,
|
| 436 |
+
"loss": 0.1415,
|
| 437 |
+
"step": 600
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 1.9551282051282053,
|
| 441 |
+
"grad_norm": 9.433862686157227,
|
| 442 |
+
"learning_rate": 1.7414529914529913e-05,
|
| 443 |
+
"loss": 0.1851,
|
| 444 |
+
"step": 610
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 1.9871794871794872,
|
| 448 |
+
"grad_norm": 0.17067989706993103,
|
| 449 |
+
"learning_rate": 1.688034188034188e-05,
|
| 450 |
+
"loss": 0.2139,
|
| 451 |
+
"step": 620
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 2.0,
|
| 455 |
+
"eval_loss": 0.295738160610199,
|
| 456 |
+
"eval_runtime": 7.5117,
|
| 457 |
+
"eval_samples_per_second": 166.141,
|
| 458 |
+
"eval_steps_per_second": 41.535,
|
| 459 |
+
"step": 624
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 2.019230769230769,
|
| 463 |
+
"grad_norm": 1.5675158500671387,
|
| 464 |
+
"learning_rate": 1.6346153846153847e-05,
|
| 465 |
+
"loss": 0.1446,
|
| 466 |
+
"step": 630
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 2.051282051282051,
|
| 470 |
+
"grad_norm": 6.23939323425293,
|
| 471 |
+
"learning_rate": 1.581196581196581e-05,
|
| 472 |
+
"loss": 0.12,
|
| 473 |
+
"step": 640
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 2.0833333333333335,
|
| 477 |
+
"grad_norm": 1.7056013345718384,
|
| 478 |
+
"learning_rate": 1.527777777777778e-05,
|
| 479 |
+
"loss": 0.0515,
|
| 480 |
+
"step": 650
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 2.1153846153846154,
|
| 484 |
+
"grad_norm": 1.1206624507904053,
|
| 485 |
+
"learning_rate": 1.4743589743589745e-05,
|
| 486 |
+
"loss": 0.0854,
|
| 487 |
+
"step": 660
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 2.1474358974358974,
|
| 491 |
+
"grad_norm": 0.023267656564712524,
|
| 492 |
+
"learning_rate": 1.4209401709401709e-05,
|
| 493 |
+
"loss": 0.1587,
|
| 494 |
+
"step": 670
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 2.1794871794871793,
|
| 498 |
+
"grad_norm": 0.08649634569883347,
|
| 499 |
+
"learning_rate": 1.3675213675213677e-05,
|
| 500 |
+
"loss": 0.0887,
|
| 501 |
+
"step": 680
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 2.2115384615384617,
|
| 505 |
+
"grad_norm": 8.077009201049805,
|
| 506 |
+
"learning_rate": 1.3141025641025642e-05,
|
| 507 |
+
"loss": 0.0607,
|
| 508 |
+
"step": 690
|
| 509 |
+
},
|
| 510 |
+
{
|
| 511 |
+
"epoch": 2.2435897435897436,
|
| 512 |
+
"grad_norm": 14.778282165527344,
|
| 513 |
+
"learning_rate": 1.2606837606837608e-05,
|
| 514 |
+
"loss": 0.1752,
|
| 515 |
+
"step": 700
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 2.2756410256410255,
|
| 519 |
+
"grad_norm": 0.2188805788755417,
|
| 520 |
+
"learning_rate": 1.2072649572649573e-05,
|
| 521 |
+
"loss": 0.1695,
|
| 522 |
+
"step": 710
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 2.3076923076923075,
|
| 526 |
+
"grad_norm": 2.394150495529175,
|
| 527 |
+
"learning_rate": 1.153846153846154e-05,
|
| 528 |
+
"loss": 0.1085,
|
| 529 |
+
"step": 720
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 2.33974358974359,
|
| 533 |
+
"grad_norm": 0.08511590212583542,
|
| 534 |
+
"learning_rate": 1.1004273504273505e-05,
|
| 535 |
+
"loss": 0.0809,
|
| 536 |
+
"step": 730
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 2.371794871794872,
|
| 540 |
+
"grad_norm": 2.596112012863159,
|
| 541 |
+
"learning_rate": 1.0470085470085471e-05,
|
| 542 |
+
"loss": 0.0928,
|
| 543 |
+
"step": 740
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 2.4038461538461537,
|
| 547 |
+
"grad_norm": 0.1449175775051117,
|
| 548 |
+
"learning_rate": 9.935897435897435e-06,
|
| 549 |
+
"loss": 0.0301,
|
| 550 |
+
"step": 750
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"epoch": 2.435897435897436,
|
| 554 |
+
"grad_norm": 6.098703861236572,
|
| 555 |
+
"learning_rate": 9.401709401709402e-06,
|
| 556 |
+
"loss": 0.1443,
|
| 557 |
+
"step": 760
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"epoch": 2.467948717948718,
|
| 561 |
+
"grad_norm": 0.07011144608259201,
|
| 562 |
+
"learning_rate": 8.867521367521369e-06,
|
| 563 |
+
"loss": 0.1343,
|
| 564 |
+
"step": 770
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 2.5,
|
| 568 |
+
"grad_norm": 0.12531624734401703,
|
| 569 |
+
"learning_rate": 8.333333333333334e-06,
|
| 570 |
+
"loss": 0.173,
|
| 571 |
+
"step": 780
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 2.532051282051282,
|
| 575 |
+
"grad_norm": 4.588987827301025,
|
| 576 |
+
"learning_rate": 7.799145299145299e-06,
|
| 577 |
+
"loss": 0.0895,
|
| 578 |
+
"step": 790
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 2.564102564102564,
|
| 582 |
+
"grad_norm": 0.29747235774993896,
|
| 583 |
+
"learning_rate": 7.264957264957266e-06,
|
| 584 |
+
"loss": 0.0675,
|
| 585 |
+
"step": 800
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 2.5961538461538463,
|
| 589 |
+
"grad_norm": 3.441330671310425,
|
| 590 |
+
"learning_rate": 6.730769230769231e-06,
|
| 591 |
+
"loss": 0.0586,
|
| 592 |
+
"step": 810
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 2.628205128205128,
|
| 596 |
+
"grad_norm": 11.991110801696777,
|
| 597 |
+
"learning_rate": 6.196581196581197e-06,
|
| 598 |
+
"loss": 0.1021,
|
| 599 |
+
"step": 820
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 2.66025641025641,
|
| 603 |
+
"grad_norm": 8.041360855102539,
|
| 604 |
+
"learning_rate": 5.662393162393163e-06,
|
| 605 |
+
"loss": 0.1902,
|
| 606 |
+
"step": 830
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"epoch": 2.6923076923076925,
|
| 610 |
+
"grad_norm": 12.715825080871582,
|
| 611 |
+
"learning_rate": 5.128205128205128e-06,
|
| 612 |
+
"loss": 0.0778,
|
| 613 |
+
"step": 840
|
| 614 |
+
},
|
| 615 |
+
{
|
| 616 |
+
"epoch": 2.7243589743589745,
|
| 617 |
+
"grad_norm": 15.647939682006836,
|
| 618 |
+
"learning_rate": 4.594017094017095e-06,
|
| 619 |
+
"loss": 0.0491,
|
| 620 |
+
"step": 850
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 2.7564102564102564,
|
| 624 |
+
"grad_norm": 0.13052542507648468,
|
| 625 |
+
"learning_rate": 4.05982905982906e-06,
|
| 626 |
+
"loss": 0.117,
|
| 627 |
+
"step": 860
|
| 628 |
+
},
|
| 629 |
+
{
|
| 630 |
+
"epoch": 2.7884615384615383,
|
| 631 |
+
"grad_norm": 6.124665260314941,
|
| 632 |
+
"learning_rate": 3.525641025641026e-06,
|
| 633 |
+
"loss": 0.0542,
|
| 634 |
+
"step": 870
|
| 635 |
+
},
|
| 636 |
+
{
|
| 637 |
+
"epoch": 2.8205128205128203,
|
| 638 |
+
"grad_norm": 10.497450828552246,
|
| 639 |
+
"learning_rate": 2.991452991452992e-06,
|
| 640 |
+
"loss": 0.0797,
|
| 641 |
+
"step": 880
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 2.8525641025641026,
|
| 645 |
+
"grad_norm": 1.3950176239013672,
|
| 646 |
+
"learning_rate": 2.4572649572649574e-06,
|
| 647 |
+
"loss": 0.0268,
|
| 648 |
+
"step": 890
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 2.8846153846153846,
|
| 652 |
+
"grad_norm": 0.22984395921230316,
|
| 653 |
+
"learning_rate": 1.9230769230769234e-06,
|
| 654 |
+
"loss": 0.1544,
|
| 655 |
+
"step": 900
|
| 656 |
+
},
|
| 657 |
+
{
|
| 658 |
+
"epoch": 2.9166666666666665,
|
| 659 |
+
"grad_norm": 4.88055944442749,
|
| 660 |
+
"learning_rate": 1.388888888888889e-06,
|
| 661 |
+
"loss": 0.1185,
|
| 662 |
+
"step": 910
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 2.948717948717949,
|
| 666 |
+
"grad_norm": 5.498613357543945,
|
| 667 |
+
"learning_rate": 8.547008547008548e-07,
|
| 668 |
+
"loss": 0.0767,
|
| 669 |
+
"step": 920
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 2.980769230769231,
|
| 673 |
+
"grad_norm": 0.36016952991485596,
|
| 674 |
+
"learning_rate": 3.205128205128205e-07,
|
| 675 |
+
"loss": 0.1011,
|
| 676 |
+
"step": 930
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 3.0,
|
| 680 |
+
"eval_loss": 0.3319796919822693,
|
| 681 |
+
"eval_runtime": 7.6953,
|
| 682 |
+
"eval_samples_per_second": 162.176,
|
| 683 |
+
"eval_steps_per_second": 40.544,
|
| 684 |
+
"step": 936
|
| 685 |
+
}
|
| 686 |
+
],
|
| 687 |
+
"logging_steps": 10,
|
| 688 |
+
"max_steps": 936,
|
| 689 |
+
"num_input_tokens_seen": 0,
|
| 690 |
+
"num_train_epochs": 3,
|
| 691 |
+
"save_steps": 100,
|
| 692 |
+
"stateful_callbacks": {
|
| 693 |
+
"TrainerControl": {
|
| 694 |
+
"args": {
|
| 695 |
+
"should_epoch_stop": false,
|
| 696 |
+
"should_evaluate": false,
|
| 697 |
+
"should_log": false,
|
| 698 |
+
"should_save": true,
|
| 699 |
+
"should_training_stop": true
|
| 700 |
+
},
|
| 701 |
+
"attributes": {}
|
| 702 |
+
}
|
| 703 |
+
},
|
| 704 |
+
"total_flos": 1349242711375872.0,
|
| 705 |
+
"train_batch_size": 4,
|
| 706 |
+
"trial_name": null,
|
| 707 |
+
"trial_params": null
|
| 708 |
+
}
|
pet-collaboration-model-facebook-bart-large-mnli_bs4_gas4_lr5e-05_ep3/checkpoint-936/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f4e314e6501027c96133414ac80199c51dfc490cecd8b84af2092dfb39cb967f
|
| 3 |
+
size 5432
|