End of full fine-tuning with BART
Browse files- checkpoint-9000/config.json +73 -0
- checkpoint-9000/generation_config.json +12 -0
- checkpoint-9000/merges.txt +0 -0
- checkpoint-9000/model.safetensors +3 -0
- checkpoint-9000/optimizer.pt +3 -0
- checkpoint-9000/rng_state.pth +3 -0
- checkpoint-9000/scheduler.pt +3 -0
- checkpoint-9000/special_tokens_map.json +15 -0
- checkpoint-9000/tokenizer.json +0 -0
- checkpoint-9000/tokenizer_config.json +58 -0
- checkpoint-9000/trainer_state.json +907 -0
- checkpoint-9000/training_args.bin +3 -0
- checkpoint-9000/vocab.json +0 -0
- checkpoint-9375/config.json +73 -0
- checkpoint-9375/generation_config.json +12 -0
- checkpoint-9375/merges.txt +0 -0
- checkpoint-9375/model.safetensors +3 -0
- checkpoint-9375/optimizer.pt +3 -0
- checkpoint-9375/rng_state.pth +3 -0
- checkpoint-9375/scheduler.pt +3 -0
- checkpoint-9375/special_tokens_map.json +15 -0
- checkpoint-9375/tokenizer.json +0 -0
- checkpoint-9375/tokenizer_config.json +58 -0
- checkpoint-9375/trainer_state.json +928 -0
- checkpoint-9375/training_args.bin +3 -0
- checkpoint-9375/vocab.json +0 -0
- config.json +73 -0
- generation_config.json +12 -0
- merges.txt +0 -0
- model.safetensors +3 -0
- special_tokens_map.json +15 -0
- tokenizer.json +0 -0
- tokenizer_config.json +58 -0
- training_args.bin +3 -0
- vocab.json +0 -0
checkpoint-9000/config.json
ADDED
|
@@ -0,0 +1,73 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"activation_dropout": 0.1,
|
| 3 |
+
"activation_function": "gelu",
|
| 4 |
+
"add_bias_logits": false,
|
| 5 |
+
"add_final_layer_norm": false,
|
| 6 |
+
"architectures": [
|
| 7 |
+
"BartForConditionalGeneration"
|
| 8 |
+
],
|
| 9 |
+
"attention_dropout": 0.1,
|
| 10 |
+
"bos_token_id": 0,
|
| 11 |
+
"classif_dropout": 0.1,
|
| 12 |
+
"classifier_dropout": 0.0,
|
| 13 |
+
"d_model": 768,
|
| 14 |
+
"decoder_attention_heads": 12,
|
| 15 |
+
"decoder_ffn_dim": 3072,
|
| 16 |
+
"decoder_layerdrop": 0.0,
|
| 17 |
+
"decoder_layers": 6,
|
| 18 |
+
"decoder_start_token_id": 2,
|
| 19 |
+
"dropout": 0.1,
|
| 20 |
+
"early_stopping": null,
|
| 21 |
+
"encoder_attention_heads": 12,
|
| 22 |
+
"encoder_ffn_dim": 3072,
|
| 23 |
+
"encoder_layerdrop": 0.0,
|
| 24 |
+
"encoder_layers": 6,
|
| 25 |
+
"eos_token_id": 2,
|
| 26 |
+
"forced_eos_token_id": 2,
|
| 27 |
+
"gradient_checkpointing": false,
|
| 28 |
+
"id2label": {
|
| 29 |
+
"0": "LABEL_0",
|
| 30 |
+
"1": "LABEL_1",
|
| 31 |
+
"2": "LABEL_2"
|
| 32 |
+
},
|
| 33 |
+
"init_std": 0.02,
|
| 34 |
+
"is_encoder_decoder": true,
|
| 35 |
+
"label2id": {
|
| 36 |
+
"LABEL_0": 0,
|
| 37 |
+
"LABEL_1": 1,
|
| 38 |
+
"LABEL_2": 2
|
| 39 |
+
},
|
| 40 |
+
"max_position_embeddings": 1024,
|
| 41 |
+
"model_type": "bart",
|
| 42 |
+
"no_repeat_ngram_size": null,
|
| 43 |
+
"normalize_before": false,
|
| 44 |
+
"normalize_embedding": true,
|
| 45 |
+
"num_beams": null,
|
| 46 |
+
"num_hidden_layers": 6,
|
| 47 |
+
"pad_token_id": 1,
|
| 48 |
+
"scale_embedding": false,
|
| 49 |
+
"task_specific_params": {
|
| 50 |
+
"summarization": {
|
| 51 |
+
"length_penalty": 1.0,
|
| 52 |
+
"max_length": 128,
|
| 53 |
+
"min_length": 12,
|
| 54 |
+
"num_beams": 4
|
| 55 |
+
},
|
| 56 |
+
"summarization_cnn": {
|
| 57 |
+
"length_penalty": 2.0,
|
| 58 |
+
"max_length": 142,
|
| 59 |
+
"min_length": 56,
|
| 60 |
+
"num_beams": 4
|
| 61 |
+
},
|
| 62 |
+
"summarization_xsum": {
|
| 63 |
+
"length_penalty": 1.0,
|
| 64 |
+
"max_length": 62,
|
| 65 |
+
"min_length": 11,
|
| 66 |
+
"num_beams": 6
|
| 67 |
+
}
|
| 68 |
+
},
|
| 69 |
+
"torch_dtype": "float32",
|
| 70 |
+
"transformers_version": "4.55.4",
|
| 71 |
+
"use_cache": true,
|
| 72 |
+
"vocab_size": 50265
|
| 73 |
+
}
|
checkpoint-9000/generation_config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 0,
|
| 3 |
+
"decoder_start_token_id": 2,
|
| 4 |
+
"early_stopping": true,
|
| 5 |
+
"eos_token_id": 2,
|
| 6 |
+
"forced_bos_token_id": 0,
|
| 7 |
+
"forced_eos_token_id": 2,
|
| 8 |
+
"no_repeat_ngram_size": 3,
|
| 9 |
+
"num_beams": 4,
|
| 10 |
+
"pad_token_id": 1,
|
| 11 |
+
"transformers_version": "4.55.4"
|
| 12 |
+
}
|
checkpoint-9000/merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-9000/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2aec4c08609cfd6d0c21bab230dba30501b0c4d262386f1cc13e06a064d4dbac
|
| 3 |
+
size 557912620
|
checkpoint-9000/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0fed046bc4f34206dff59e0457c6f0eb3530b9698a4a4acf6350ce7bdc3e5cf5
|
| 3 |
+
size 1115579898
|
checkpoint-9000/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cb09790d4971a40e0836de21e1ad3d7be3f82c74ecfe27dcd9231598562191c0
|
| 3 |
+
size 14244
|
checkpoint-9000/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a6cc646667b5c88e34e6af3d7510104c2148b37d14a63725b931f502cadad7fb
|
| 3 |
+
size 1064
|
checkpoint-9000/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": true,
|
| 9 |
+
"rstrip": false,
|
| 10 |
+
"single_word": false
|
| 11 |
+
},
|
| 12 |
+
"pad_token": "<pad>",
|
| 13 |
+
"sep_token": "</s>",
|
| 14 |
+
"unk_token": "<unk>"
|
| 15 |
+
}
|
checkpoint-9000/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-9000/tokenizer_config.json
ADDED
|
@@ -0,0 +1,58 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"added_tokens_decoder": {
|
| 4 |
+
"0": {
|
| 5 |
+
"content": "<s>",
|
| 6 |
+
"lstrip": false,
|
| 7 |
+
"normalized": true,
|
| 8 |
+
"rstrip": false,
|
| 9 |
+
"single_word": false,
|
| 10 |
+
"special": true
|
| 11 |
+
},
|
| 12 |
+
"1": {
|
| 13 |
+
"content": "<pad>",
|
| 14 |
+
"lstrip": false,
|
| 15 |
+
"normalized": true,
|
| 16 |
+
"rstrip": false,
|
| 17 |
+
"single_word": false,
|
| 18 |
+
"special": true
|
| 19 |
+
},
|
| 20 |
+
"2": {
|
| 21 |
+
"content": "</s>",
|
| 22 |
+
"lstrip": false,
|
| 23 |
+
"normalized": true,
|
| 24 |
+
"rstrip": false,
|
| 25 |
+
"single_word": false,
|
| 26 |
+
"special": true
|
| 27 |
+
},
|
| 28 |
+
"3": {
|
| 29 |
+
"content": "<unk>",
|
| 30 |
+
"lstrip": false,
|
| 31 |
+
"normalized": true,
|
| 32 |
+
"rstrip": false,
|
| 33 |
+
"single_word": false,
|
| 34 |
+
"special": true
|
| 35 |
+
},
|
| 36 |
+
"50264": {
|
| 37 |
+
"content": "<mask>",
|
| 38 |
+
"lstrip": true,
|
| 39 |
+
"normalized": true,
|
| 40 |
+
"rstrip": false,
|
| 41 |
+
"single_word": false,
|
| 42 |
+
"special": true
|
| 43 |
+
}
|
| 44 |
+
},
|
| 45 |
+
"bos_token": "<s>",
|
| 46 |
+
"clean_up_tokenization_spaces": false,
|
| 47 |
+
"cls_token": "<s>",
|
| 48 |
+
"eos_token": "</s>",
|
| 49 |
+
"errors": "replace",
|
| 50 |
+
"extra_special_tokens": {},
|
| 51 |
+
"mask_token": "<mask>",
|
| 52 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 53 |
+
"pad_token": "<pad>",
|
| 54 |
+
"sep_token": "</s>",
|
| 55 |
+
"tokenizer_class": "BartTokenizer",
|
| 56 |
+
"trim_offsets": true,
|
| 57 |
+
"unk_token": "<unk>"
|
| 58 |
+
}
|
checkpoint-9000/trainer_state.json
ADDED
|
@@ -0,0 +1,907 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 9000,
|
| 3 |
+
"best_metric": 0.009878098033368587,
|
| 4 |
+
"best_model_checkpoint": "bart-base-finetune/checkpoint-9000",
|
| 5 |
+
"epoch": 2.88,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 9000,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.032,
|
| 14 |
+
"grad_norm": 3.6718413829803467,
|
| 15 |
+
"learning_rate": 9.900000000000002e-06,
|
| 16 |
+
"loss": 2.4853,
|
| 17 |
+
"step": 100
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.064,
|
| 21 |
+
"grad_norm": 2.339207649230957,
|
| 22 |
+
"learning_rate": 1.9900000000000003e-05,
|
| 23 |
+
"loss": 0.5438,
|
| 24 |
+
"step": 200
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.096,
|
| 28 |
+
"grad_norm": 1.8246095180511475,
|
| 29 |
+
"learning_rate": 2.9900000000000002e-05,
|
| 30 |
+
"loss": 0.3038,
|
| 31 |
+
"step": 300
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.128,
|
| 35 |
+
"grad_norm": 1.9325376749038696,
|
| 36 |
+
"learning_rate": 3.99e-05,
|
| 37 |
+
"loss": 0.2137,
|
| 38 |
+
"step": 400
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.16,
|
| 42 |
+
"grad_norm": 1.487189531326294,
|
| 43 |
+
"learning_rate": 4.99e-05,
|
| 44 |
+
"loss": 0.1689,
|
| 45 |
+
"step": 500
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.16,
|
| 49 |
+
"eval_gen_len": 20.9688,
|
| 50 |
+
"eval_loss": 0.10288307070732117,
|
| 51 |
+
"eval_rouge1": 0.4964,
|
| 52 |
+
"eval_rouge2": 0.4123,
|
| 53 |
+
"eval_rougeL": 0.4927,
|
| 54 |
+
"eval_rougeLsum": 0.4927,
|
| 55 |
+
"eval_runtime": 144.5729,
|
| 56 |
+
"eval_samples_per_second": 34.585,
|
| 57 |
+
"eval_steps_per_second": 2.165,
|
| 58 |
+
"step": 500
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.192,
|
| 62 |
+
"grad_norm": 0.9393866062164307,
|
| 63 |
+
"learning_rate": 4.998465033232545e-05,
|
| 64 |
+
"loss": 0.1361,
|
| 65 |
+
"step": 600
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.224,
|
| 69 |
+
"grad_norm": 1.4453977346420288,
|
| 70 |
+
"learning_rate": 4.9937998871479073e-05,
|
| 71 |
+
"loss": 0.1156,
|
| 72 |
+
"step": 700
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.256,
|
| 76 |
+
"grad_norm": 1.1997668743133545,
|
| 77 |
+
"learning_rate": 4.9860102500953126e-05,
|
| 78 |
+
"loss": 0.0988,
|
| 79 |
+
"step": 800
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.288,
|
| 83 |
+
"grad_norm": 1.3937125205993652,
|
| 84 |
+
"learning_rate": 4.975105881739415e-05,
|
| 85 |
+
"loss": 0.0889,
|
| 86 |
+
"step": 900
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.32,
|
| 90 |
+
"grad_norm": 1.3387874364852905,
|
| 91 |
+
"learning_rate": 4.961100444203071e-05,
|
| 92 |
+
"loss": 0.0793,
|
| 93 |
+
"step": 1000
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.32,
|
| 97 |
+
"eval_gen_len": 20.9698,
|
| 98 |
+
"eval_loss": 0.05177076905965805,
|
| 99 |
+
"eval_rouge1": 0.4988,
|
| 100 |
+
"eval_rouge2": 0.4239,
|
| 101 |
+
"eval_rougeL": 0.494,
|
| 102 |
+
"eval_rougeLsum": 0.494,
|
| 103 |
+
"eval_runtime": 144.6374,
|
| 104 |
+
"eval_samples_per_second": 34.569,
|
| 105 |
+
"eval_steps_per_second": 2.164,
|
| 106 |
+
"step": 1000
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 0.352,
|
| 110 |
+
"grad_norm": 1.226305365562439,
|
| 111 |
+
"learning_rate": 4.944011484950018e-05,
|
| 112 |
+
"loss": 0.0726,
|
| 113 |
+
"step": 1100
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 0.384,
|
| 117 |
+
"grad_norm": 1.3799694776535034,
|
| 118 |
+
"learning_rate": 4.9238604147995884e-05,
|
| 119 |
+
"loss": 0.0708,
|
| 120 |
+
"step": 1200
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 0.416,
|
| 124 |
+
"grad_norm": 1.0327191352844238,
|
| 125 |
+
"learning_rate": 4.900672481101019e-05,
|
| 126 |
+
"loss": 0.0675,
|
| 127 |
+
"step": 1300
|
| 128 |
+
},
|
| 129 |
+
{
|
| 130 |
+
"epoch": 0.448,
|
| 131 |
+
"grad_norm": 0.8962351083755493,
|
| 132 |
+
"learning_rate": 4.8744767361009594e-05,
|
| 133 |
+
"loss": 0.0604,
|
| 134 |
+
"step": 1400
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 0.48,
|
| 138 |
+
"grad_norm": 0.6496387720108032,
|
| 139 |
+
"learning_rate": 4.8453060005437934e-05,
|
| 140 |
+
"loss": 0.0568,
|
| 141 |
+
"step": 1500
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 0.48,
|
| 145 |
+
"eval_gen_len": 20.9742,
|
| 146 |
+
"eval_loss": 0.03353753313422203,
|
| 147 |
+
"eval_rouge1": 0.5059,
|
| 148 |
+
"eval_rouge2": 0.4356,
|
| 149 |
+
"eval_rougeL": 0.5032,
|
| 150 |
+
"eval_rougeLsum": 0.503,
|
| 151 |
+
"eval_runtime": 145.5239,
|
| 152 |
+
"eval_samples_per_second": 34.359,
|
| 153 |
+
"eval_steps_per_second": 2.151,
|
| 154 |
+
"step": 1500
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 0.512,
|
| 158 |
+
"grad_norm": 1.6739161014556885,
|
| 159 |
+
"learning_rate": 4.8131968225504245e-05,
|
| 160 |
+
"loss": 0.0539,
|
| 161 |
+
"step": 1600
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 0.544,
|
| 165 |
+
"grad_norm": 1.063624382019043,
|
| 166 |
+
"learning_rate": 4.778189431826991e-05,
|
| 167 |
+
"loss": 0.0502,
|
| 168 |
+
"step": 1700
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 0.576,
|
| 172 |
+
"grad_norm": 1.1451120376586914,
|
| 173 |
+
"learning_rate": 4.7403276892609224e-05,
|
| 174 |
+
"loss": 0.0477,
|
| 175 |
+
"step": 1800
|
| 176 |
+
},
|
| 177 |
+
{
|
| 178 |
+
"epoch": 0.608,
|
| 179 |
+
"grad_norm": 0.7193688750267029,
|
| 180 |
+
"learning_rate": 4.699659031967479e-05,
|
| 181 |
+
"loss": 0.0457,
|
| 182 |
+
"step": 1900
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 0.64,
|
| 186 |
+
"grad_norm": 0.9038898944854736,
|
| 187 |
+
"learning_rate": 4.656234413855613e-05,
|
| 188 |
+
"loss": 0.0434,
|
| 189 |
+
"step": 2000
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 0.64,
|
| 193 |
+
"eval_gen_len": 20.9714,
|
| 194 |
+
"eval_loss": 0.029211629182100296,
|
| 195 |
+
"eval_rouge1": 0.5076,
|
| 196 |
+
"eval_rouge2": 0.4388,
|
| 197 |
+
"eval_rougeL": 0.5055,
|
| 198 |
+
"eval_rougeLsum": 0.5054,
|
| 199 |
+
"eval_runtime": 145.669,
|
| 200 |
+
"eval_samples_per_second": 34.324,
|
| 201 |
+
"eval_steps_per_second": 2.149,
|
| 202 |
+
"step": 2000
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 0.672,
|
| 206 |
+
"grad_norm": 1.7222179174423218,
|
| 207 |
+
"learning_rate": 4.610108241787629e-05,
|
| 208 |
+
"loss": 0.0461,
|
| 209 |
+
"step": 2100
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 0.704,
|
| 213 |
+
"grad_norm": 0.6831910014152527,
|
| 214 |
+
"learning_rate": 4.561338307412629e-05,
|
| 215 |
+
"loss": 0.0438,
|
| 216 |
+
"step": 2200
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 0.736,
|
| 220 |
+
"grad_norm": 0.4437102675437927,
|
| 221 |
+
"learning_rate": 4.509985714759151e-05,
|
| 222 |
+
"loss": 0.0407,
|
| 223 |
+
"step": 2300
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"epoch": 0.768,
|
| 227 |
+
"grad_norm": 0.6408787965774536,
|
| 228 |
+
"learning_rate": 4.456114803677696e-05,
|
| 229 |
+
"loss": 0.0377,
|
| 230 |
+
"step": 2400
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 0.8,
|
| 234 |
+
"grad_norm": 0.7488101124763489,
|
| 235 |
+
"learning_rate": 4.3997930692291056e-05,
|
| 236 |
+
"loss": 0.0372,
|
| 237 |
+
"step": 2500
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 0.8,
|
| 241 |
+
"eval_gen_len": 20.9734,
|
| 242 |
+
"eval_loss": 0.02469678223133087,
|
| 243 |
+
"eval_rouge1": 0.5089,
|
| 244 |
+
"eval_rouge2": 0.4398,
|
| 245 |
+
"eval_rougeL": 0.5063,
|
| 246 |
+
"eval_rougeLsum": 0.5061,
|
| 247 |
+
"eval_runtime": 145.4374,
|
| 248 |
+
"eval_samples_per_second": 34.379,
|
| 249 |
+
"eval_steps_per_second": 2.152,
|
| 250 |
+
"step": 2500
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 0.832,
|
| 254 |
+
"grad_norm": 0.3353654742240906,
|
| 255 |
+
"learning_rate": 4.341091077119753e-05,
|
| 256 |
+
"loss": 0.0367,
|
| 257 |
+
"step": 2600
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 0.864,
|
| 261 |
+
"grad_norm": 0.853186845779419,
|
| 262 |
+
"learning_rate": 4.2800823752895055e-05,
|
| 263 |
+
"loss": 0.0345,
|
| 264 |
+
"step": 2700
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 0.896,
|
| 268 |
+
"grad_norm": 0.7370714545249939,
|
| 269 |
+
"learning_rate": 4.2168434017632516e-05,
|
| 270 |
+
"loss": 0.038,
|
| 271 |
+
"step": 2800
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 0.928,
|
| 275 |
+
"grad_norm": 0.5349611639976501,
|
| 276 |
+
"learning_rate": 4.151453388881413e-05,
|
| 277 |
+
"loss": 0.0321,
|
| 278 |
+
"step": 2900
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 0.96,
|
| 282 |
+
"grad_norm": 0.7140136957168579,
|
| 283 |
+
"learning_rate": 4.083994264029462e-05,
|
| 284 |
+
"loss": 0.0347,
|
| 285 |
+
"step": 3000
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 0.96,
|
| 289 |
+
"eval_gen_len": 20.9754,
|
| 290 |
+
"eval_loss": 0.01928527094423771,
|
| 291 |
+
"eval_rouge1": 0.5082,
|
| 292 |
+
"eval_rouge2": 0.4406,
|
| 293 |
+
"eval_rougeL": 0.5057,
|
| 294 |
+
"eval_rougeLsum": 0.5055,
|
| 295 |
+
"eval_runtime": 145.567,
|
| 296 |
+
"eval_samples_per_second": 34.348,
|
| 297 |
+
"eval_steps_per_second": 2.15,
|
| 298 |
+
"step": 3000
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"epoch": 0.992,
|
| 302 |
+
"grad_norm": 0.6110441088676453,
|
| 303 |
+
"learning_rate": 4.0145505469907984e-05,
|
| 304 |
+
"loss": 0.0325,
|
| 305 |
+
"step": 3100
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 1.024,
|
| 309 |
+
"grad_norm": 0.4845236539840698,
|
| 310 |
+
"learning_rate": 3.9432092440516075e-05,
|
| 311 |
+
"loss": 0.0276,
|
| 312 |
+
"step": 3200
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 1.056,
|
| 316 |
+
"grad_norm": 0.5893353819847107,
|
| 317 |
+
"learning_rate": 3.8700597389903664e-05,
|
| 318 |
+
"loss": 0.0262,
|
| 319 |
+
"step": 3300
|
| 320 |
+
},
|
| 321 |
+
{
|
| 322 |
+
"epoch": 1.088,
|
| 323 |
+
"grad_norm": 0.7486633658409119,
|
| 324 |
+
"learning_rate": 3.795193681088578e-05,
|
| 325 |
+
"loss": 0.0235,
|
| 326 |
+
"step": 3400
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 1.12,
|
| 330 |
+
"grad_norm": 0.886471688747406,
|
| 331 |
+
"learning_rate": 3.7187048703030597e-05,
|
| 332 |
+
"loss": 0.0235,
|
| 333 |
+
"step": 3500
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 1.12,
|
| 337 |
+
"eval_gen_len": 20.9758,
|
| 338 |
+
"eval_loss": 0.0190653745085001,
|
| 339 |
+
"eval_rouge1": 0.5087,
|
| 340 |
+
"eval_rouge2": 0.4431,
|
| 341 |
+
"eval_rougeL": 0.5072,
|
| 342 |
+
"eval_rougeLsum": 0.5071,
|
| 343 |
+
"eval_runtime": 145.2595,
|
| 344 |
+
"eval_samples_per_second": 34.421,
|
| 345 |
+
"eval_steps_per_second": 2.155,
|
| 346 |
+
"step": 3500
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 1.152,
|
| 350 |
+
"grad_norm": 1.1492247581481934,
|
| 351 |
+
"learning_rate": 3.640689139743628e-05,
|
| 352 |
+
"loss": 0.0234,
|
| 353 |
+
"step": 3600
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 1.184,
|
| 357 |
+
"grad_norm": 0.7707476019859314,
|
| 358 |
+
"learning_rate": 3.561244235603451e-05,
|
| 359 |
+
"loss": 0.0245,
|
| 360 |
+
"step": 3700
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 1.216,
|
| 364 |
+
"grad_norm": 0.8048288822174072,
|
| 365 |
+
"learning_rate": 3.4804696946924855e-05,
|
| 366 |
+
"loss": 0.0228,
|
| 367 |
+
"step": 3800
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 1.248,
|
| 371 |
+
"grad_norm": 0.7920108437538147,
|
| 372 |
+
"learning_rate": 3.398466719727445e-05,
|
| 373 |
+
"loss": 0.0227,
|
| 374 |
+
"step": 3900
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 1.28,
|
| 378 |
+
"grad_norm": 0.199005588889122,
|
| 379 |
+
"learning_rate": 3.3153380525345445e-05,
|
| 380 |
+
"loss": 0.0226,
|
| 381 |
+
"step": 4000
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 1.28,
|
| 385 |
+
"eval_gen_len": 20.975,
|
| 386 |
+
"eval_loss": 0.0172447320073843,
|
| 387 |
+
"eval_rouge1": 0.5091,
|
| 388 |
+
"eval_rouge2": 0.442,
|
| 389 |
+
"eval_rougeL": 0.5064,
|
| 390 |
+
"eval_rougeLsum": 0.5063,
|
| 391 |
+
"eval_runtime": 145.4831,
|
| 392 |
+
"eval_samples_per_second": 34.368,
|
| 393 |
+
"eval_steps_per_second": 2.151,
|
| 394 |
+
"step": 4000
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 1.312,
|
| 398 |
+
"grad_norm": 0.6734442114830017,
|
| 399 |
+
"learning_rate": 3.2311878453238966e-05,
|
| 400 |
+
"loss": 0.0207,
|
| 401 |
+
"step": 4100
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 1.3439999999999999,
|
| 405 |
+
"grad_norm": 0.5161114931106567,
|
| 406 |
+
"learning_rate": 3.146121530196824e-05,
|
| 407 |
+
"loss": 0.0216,
|
| 408 |
+
"step": 4200
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 1.376,
|
| 412 |
+
"grad_norm": 0.4449058175086975,
|
| 413 |
+
"learning_rate": 3.0602456870495973e-05,
|
| 414 |
+
"loss": 0.0197,
|
| 415 |
+
"step": 4300
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 1.408,
|
| 419 |
+
"grad_norm": 0.5729345083236694,
|
| 420 |
+
"learning_rate": 2.9736679100390947e-05,
|
| 421 |
+
"loss": 0.0208,
|
| 422 |
+
"step": 4400
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 1.44,
|
| 426 |
+
"grad_norm": 0.19021901488304138,
|
| 427 |
+
"learning_rate": 2.886496672777688e-05,
|
| 428 |
+
"loss": 0.0201,
|
| 429 |
+
"step": 4500
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 1.44,
|
| 433 |
+
"eval_gen_len": 20.9758,
|
| 434 |
+
"eval_loss": 0.014488083310425282,
|
| 435 |
+
"eval_rouge1": 0.5116,
|
| 436 |
+
"eval_rouge2": 0.4474,
|
| 437 |
+
"eval_rougeL": 0.5102,
|
| 438 |
+
"eval_rougeLsum": 0.5101,
|
| 439 |
+
"eval_runtime": 146.2423,
|
| 440 |
+
"eval_samples_per_second": 34.19,
|
| 441 |
+
"eval_steps_per_second": 2.14,
|
| 442 |
+
"step": 4500
|
| 443 |
+
},
|
| 444 |
+
{
|
| 445 |
+
"epoch": 1.472,
|
| 446 |
+
"grad_norm": 0.632882297039032,
|
| 447 |
+
"learning_rate": 2.7988411924262648e-05,
|
| 448 |
+
"loss": 0.0198,
|
| 449 |
+
"step": 4600
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 1.504,
|
| 453 |
+
"grad_norm": 0.37011951208114624,
|
| 454 |
+
"learning_rate": 2.71081129285564e-05,
|
| 455 |
+
"loss": 0.0205,
|
| 456 |
+
"step": 4700
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 1.536,
|
| 460 |
+
"grad_norm": 0.528691828250885,
|
| 461 |
+
"learning_rate": 2.622517267047835e-05,
|
| 462 |
+
"loss": 0.0204,
|
| 463 |
+
"step": 4800
|
| 464 |
+
},
|
| 465 |
+
{
|
| 466 |
+
"epoch": 1.568,
|
| 467 |
+
"grad_norm": 0.19730792939662933,
|
| 468 |
+
"learning_rate": 2.5340697389095973e-05,
|
| 469 |
+
"loss": 0.0193,
|
| 470 |
+
"step": 4900
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 1.6,
|
| 474 |
+
"grad_norm": 0.5517034530639648,
|
| 475 |
+
"learning_rate": 2.4455795246713044e-05,
|
| 476 |
+
"loss": 0.0164,
|
| 477 |
+
"step": 5000
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 1.6,
|
| 481 |
+
"eval_gen_len": 20.9766,
|
| 482 |
+
"eval_loss": 0.013988671824336052,
|
| 483 |
+
"eval_rouge1": 0.5114,
|
| 484 |
+
"eval_rouge2": 0.4467,
|
| 485 |
+
"eval_rougeL": 0.5099,
|
| 486 |
+
"eval_rougeLsum": 0.5098,
|
| 487 |
+
"eval_runtime": 145.249,
|
| 488 |
+
"eval_samples_per_second": 34.424,
|
| 489 |
+
"eval_steps_per_second": 2.155,
|
| 490 |
+
"step": 5000
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 1.6320000000000001,
|
| 494 |
+
"grad_norm": 0.6311265826225281,
|
| 495 |
+
"learning_rate": 2.3571574940449008e-05,
|
| 496 |
+
"loss": 0.0173,
|
| 497 |
+
"step": 5100
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 1.6640000000000001,
|
| 501 |
+
"grad_norm": 0.764806866645813,
|
| 502 |
+
"learning_rate": 2.2689144313148382e-05,
|
| 503 |
+
"loss": 0.0173,
|
| 504 |
+
"step": 5200
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 1.696,
|
| 508 |
+
"grad_norm": 1.1502550840377808,
|
| 509 |
+
"learning_rate": 2.180960896536036e-05,
|
| 510 |
+
"loss": 0.018,
|
| 511 |
+
"step": 5300
|
| 512 |
+
},
|
| 513 |
+
{
|
| 514 |
+
"epoch": 1.728,
|
| 515 |
+
"grad_norm": 0.21649418771266937,
|
| 516 |
+
"learning_rate": 2.0934070870127912e-05,
|
| 517 |
+
"loss": 0.0162,
|
| 518 |
+
"step": 5400
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 1.76,
|
| 522 |
+
"grad_norm": 0.3456017076969147,
|
| 523 |
+
"learning_rate": 2.0063626992321693e-05,
|
| 524 |
+
"loss": 0.017,
|
| 525 |
+
"step": 5500
|
| 526 |
+
},
|
| 527 |
+
{
|
| 528 |
+
"epoch": 1.76,
|
| 529 |
+
"eval_gen_len": 20.9772,
|
| 530 |
+
"eval_loss": 0.012542162090539932,
|
| 531 |
+
"eval_rouge1": 0.5119,
|
| 532 |
+
"eval_rouge2": 0.4484,
|
| 533 |
+
"eval_rougeL": 0.5108,
|
| 534 |
+
"eval_rougeLsum": 0.5106,
|
| 535 |
+
"eval_runtime": 144.5364,
|
| 536 |
+
"eval_samples_per_second": 34.593,
|
| 537 |
+
"eval_steps_per_second": 2.166,
|
| 538 |
+
"step": 5500
|
| 539 |
+
},
|
| 540 |
+
{
|
| 541 |
+
"epoch": 1.792,
|
| 542 |
+
"grad_norm": 0.5721284747123718,
|
| 543 |
+
"learning_rate": 1.919936791424879e-05,
|
| 544 |
+
"loss": 0.0176,
|
| 545 |
+
"step": 5600
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 1.8239999999999998,
|
| 549 |
+
"grad_norm": 0.26220762729644775,
|
| 550 |
+
"learning_rate": 1.834237646925815e-05,
|
| 551 |
+
"loss": 0.0152,
|
| 552 |
+
"step": 5700
|
| 553 |
+
},
|
| 554 |
+
{
|
| 555 |
+
"epoch": 1.8559999999999999,
|
| 556 |
+
"grad_norm": 0.271634966135025,
|
| 557 |
+
"learning_rate": 1.7493726385054743e-05,
|
| 558 |
+
"loss": 0.0152,
|
| 559 |
+
"step": 5800
|
| 560 |
+
},
|
| 561 |
+
{
|
| 562 |
+
"epoch": 1.888,
|
| 563 |
+
"grad_norm": 0.3826049566268921,
|
| 564 |
+
"learning_rate": 1.6654480938422208e-05,
|
| 565 |
+
"loss": 0.0151,
|
| 566 |
+
"step": 5900
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 1.92,
|
| 570 |
+
"grad_norm": 0.49469685554504395,
|
| 571 |
+
"learning_rate": 1.58256916230395e-05,
|
| 572 |
+
"loss": 0.0146,
|
| 573 |
+
"step": 6000
|
| 574 |
+
},
|
| 575 |
+
{
|
| 576 |
+
"epoch": 1.92,
|
| 577 |
+
"eval_gen_len": 20.976,
|
| 578 |
+
"eval_loss": 0.011809424497187138,
|
| 579 |
+
"eval_rouge1": 0.5115,
|
| 580 |
+
"eval_rouge2": 0.4481,
|
| 581 |
+
"eval_rougeL": 0.5103,
|
| 582 |
+
"eval_rougeLsum": 0.5101,
|
| 583 |
+
"eval_runtime": 144.9109,
|
| 584 |
+
"eval_samples_per_second": 34.504,
|
| 585 |
+
"eval_steps_per_second": 2.16,
|
| 586 |
+
"step": 6000
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"epoch": 1.952,
|
| 590 |
+
"grad_norm": 0.6852744221687317,
|
| 591 |
+
"learning_rate": 1.5008396832060606e-05,
|
| 592 |
+
"loss": 0.0161,
|
| 593 |
+
"step": 6100
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 1.984,
|
| 597 |
+
"grad_norm": 0.27392491698265076,
|
| 598 |
+
"learning_rate": 1.4203620557108022e-05,
|
| 599 |
+
"loss": 0.0146,
|
| 600 |
+
"step": 6200
|
| 601 |
+
},
|
| 602 |
+
{
|
| 603 |
+
"epoch": 2.016,
|
| 604 |
+
"grad_norm": 0.6940794587135315,
|
| 605 |
+
"learning_rate": 1.3412371105309856e-05,
|
| 606 |
+
"loss": 0.0132,
|
| 607 |
+
"step": 6300
|
| 608 |
+
},
|
| 609 |
+
{
|
| 610 |
+
"epoch": 2.048,
|
| 611 |
+
"grad_norm": 0.2785722613334656,
|
| 612 |
+
"learning_rate": 1.2635639835988195e-05,
|
| 613 |
+
"loss": 0.0106,
|
| 614 |
+
"step": 6400
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 2.08,
|
| 618 |
+
"grad_norm": 0.1548278033733368,
|
| 619 |
+
"learning_rate": 1.1874399918581388e-05,
|
| 620 |
+
"loss": 0.0114,
|
| 621 |
+
"step": 6500
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"epoch": 2.08,
|
| 625 |
+
"eval_gen_len": 20.9772,
|
| 626 |
+
"eval_loss": 0.010936998762190342,
|
| 627 |
+
"eval_rouge1": 0.5117,
|
| 628 |
+
"eval_rouge2": 0.4484,
|
| 629 |
+
"eval_rougeL": 0.5105,
|
| 630 |
+
"eval_rougeLsum": 0.5105,
|
| 631 |
+
"eval_runtime": 145.1325,
|
| 632 |
+
"eval_samples_per_second": 34.451,
|
| 633 |
+
"eval_steps_per_second": 2.157,
|
| 634 |
+
"step": 6500
|
| 635 |
+
},
|
| 636 |
+
{
|
| 637 |
+
"epoch": 2.112,
|
| 638 |
+
"grad_norm": 0.14601846039295197,
|
| 639 |
+
"learning_rate": 1.1129605113356508e-05,
|
| 640 |
+
"loss": 0.0111,
|
| 641 |
+
"step": 6600
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 2.144,
|
| 645 |
+
"grad_norm": 0.18163363635540009,
|
| 646 |
+
"learning_rate": 1.040218857643962e-05,
|
| 647 |
+
"loss": 0.0111,
|
| 648 |
+
"step": 6700
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 2.176,
|
| 652 |
+
"grad_norm": 0.2548365294933319,
|
| 653 |
+
"learning_rate": 9.693061690661118e-06,
|
| 654 |
+
"loss": 0.011,
|
| 655 |
+
"step": 6800
|
| 656 |
+
},
|
| 657 |
+
{
|
| 658 |
+
"epoch": 2.208,
|
| 659 |
+
"grad_norm": 0.21669094264507294,
|
| 660 |
+
"learning_rate": 9.00311292368079e-06,
|
| 661 |
+
"loss": 0.0111,
|
| 662 |
+
"step": 6900
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 2.24,
|
| 666 |
+
"grad_norm": 0.47503963112831116,
|
| 667 |
+
"learning_rate": 8.333206714823513e-06,
|
| 668 |
+
"loss": 0.0104,
|
| 669 |
+
"step": 7000
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 2.24,
|
| 673 |
+
"eval_gen_len": 20.9772,
|
| 674 |
+
"eval_loss": 0.01037654373794794,
|
| 675 |
+
"eval_rouge1": 0.5129,
|
| 676 |
+
"eval_rouge2": 0.4501,
|
| 677 |
+
"eval_rougeL": 0.5118,
|
| 678 |
+
"eval_rougeLsum": 0.5117,
|
| 679 |
+
"eval_runtime": 143.5577,
|
| 680 |
+
"eval_samples_per_second": 34.829,
|
| 681 |
+
"eval_steps_per_second": 2.18,
|
| 682 |
+
"step": 7000
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 2.2720000000000002,
|
| 686 |
+
"grad_norm": 0.7779979705810547,
|
| 687 |
+
"learning_rate": 7.684182392020015e-06,
|
| 688 |
+
"loss": 0.0109,
|
| 689 |
+
"step": 7100
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 2.304,
|
| 693 |
+
"grad_norm": 0.7133929133415222,
|
| 694 |
+
"learning_rate": 7.056853120209888e-06,
|
| 695 |
+
"loss": 0.0117,
|
| 696 |
+
"step": 7200
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 2.336,
|
| 700 |
+
"grad_norm": 0.5449277758598328,
|
| 701 |
+
"learning_rate": 6.45200488252426e-06,
|
| 702 |
+
"loss": 0.0097,
|
| 703 |
+
"step": 7300
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 2.368,
|
| 707 |
+
"grad_norm": 0.5456063747406006,
|
| 708 |
+
"learning_rate": 5.870395495524664e-06,
|
| 709 |
+
"loss": 0.0095,
|
| 710 |
+
"step": 7400
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 2.4,
|
| 714 |
+
"grad_norm": 0.21037012338638306,
|
| 715 |
+
"learning_rate": 5.312753659731909e-06,
|
| 716 |
+
"loss": 0.01,
|
| 717 |
+
"step": 7500
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 2.4,
|
| 721 |
+
"eval_gen_len": 20.9772,
|
| 722 |
+
"eval_loss": 0.010514349676668644,
|
| 723 |
+
"eval_rouge1": 0.5124,
|
| 724 |
+
"eval_rouge2": 0.4493,
|
| 725 |
+
"eval_rougeL": 0.5111,
|
| 726 |
+
"eval_rougeLsum": 0.5111,
|
| 727 |
+
"eval_runtime": 143.501,
|
| 728 |
+
"eval_samples_per_second": 34.843,
|
| 729 |
+
"eval_steps_per_second": 2.181,
|
| 730 |
+
"step": 7500
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 2.432,
|
| 734 |
+
"grad_norm": 0.5376931428909302,
|
| 735 |
+
"learning_rate": 4.779778046634537e-06,
|
| 736 |
+
"loss": 0.0114,
|
| 737 |
+
"step": 7600
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 2.464,
|
| 741 |
+
"grad_norm": 0.655579149723053,
|
| 742 |
+
"learning_rate": 4.272136423320786e-06,
|
| 743 |
+
"loss": 0.0099,
|
| 744 |
+
"step": 7700
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 2.496,
|
| 748 |
+
"grad_norm": 0.24749213457107544,
|
| 749 |
+
"learning_rate": 3.7904648158307465e-06,
|
| 750 |
+
"loss": 0.0088,
|
| 751 |
+
"step": 7800
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 2.528,
|
| 755 |
+
"grad_norm": 0.09235558658838272,
|
| 756 |
+
"learning_rate": 3.335366712277019e-06,
|
| 757 |
+
"loss": 0.0092,
|
| 758 |
+
"step": 7900
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 2.56,
|
| 762 |
+
"grad_norm": 0.4215800166130066,
|
| 763 |
+
"learning_rate": 2.907412306732288e-06,
|
| 764 |
+
"loss": 0.01,
|
| 765 |
+
"step": 8000
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 2.56,
|
| 769 |
+
"eval_gen_len": 20.9772,
|
| 770 |
+
"eval_loss": 0.009960692375898361,
|
| 771 |
+
"eval_rouge1": 0.5131,
|
| 772 |
+
"eval_rouge2": 0.4506,
|
| 773 |
+
"eval_rougeL": 0.5119,
|
| 774 |
+
"eval_rougeLsum": 0.5119,
|
| 775 |
+
"eval_runtime": 143.2943,
|
| 776 |
+
"eval_samples_per_second": 34.893,
|
| 777 |
+
"eval_steps_per_second": 2.184,
|
| 778 |
+
"step": 8000
|
| 779 |
+
},
|
| 780 |
+
{
|
| 781 |
+
"epoch": 2.592,
|
| 782 |
+
"grad_norm": 0.15820010006427765,
|
| 783 |
+
"learning_rate": 2.5071377848310595e-06,
|
| 784 |
+
"loss": 0.0092,
|
| 785 |
+
"step": 8100
|
| 786 |
+
},
|
| 787 |
+
{
|
| 788 |
+
"epoch": 2.624,
|
| 789 |
+
"grad_norm": 0.24991130828857422,
|
| 790 |
+
"learning_rate": 2.135044651980733e-06,
|
| 791 |
+
"loss": 0.0089,
|
| 792 |
+
"step": 8200
|
| 793 |
+
},
|
| 794 |
+
{
|
| 795 |
+
"epoch": 2.656,
|
| 796 |
+
"grad_norm": 0.5366423726081848,
|
| 797 |
+
"learning_rate": 1.7915991050236502e-06,
|
| 798 |
+
"loss": 0.0099,
|
| 799 |
+
"step": 8300
|
| 800 |
+
},
|
| 801 |
+
{
|
| 802 |
+
"epoch": 2.6879999999999997,
|
| 803 |
+
"grad_norm": 0.2854973375797272,
|
| 804 |
+
"learning_rate": 1.4772314481373838e-06,
|
| 805 |
+
"loss": 0.0096,
|
| 806 |
+
"step": 8400
|
| 807 |
+
},
|
| 808 |
+
{
|
| 809 |
+
"epoch": 2.7199999999999998,
|
| 810 |
+
"grad_norm": 0.24889616668224335,
|
| 811 |
+
"learning_rate": 1.1923355537050458e-06,
|
| 812 |
+
"loss": 0.0084,
|
| 813 |
+
"step": 8500
|
| 814 |
+
},
|
| 815 |
+
{
|
| 816 |
+
"epoch": 2.7199999999999998,
|
| 817 |
+
"eval_gen_len": 20.9772,
|
| 818 |
+
"eval_loss": 0.009925846941769123,
|
| 819 |
+
"eval_rouge1": 0.5132,
|
| 820 |
+
"eval_rouge2": 0.4505,
|
| 821 |
+
"eval_rougeL": 0.512,
|
| 822 |
+
"eval_rougeLsum": 0.5119,
|
| 823 |
+
"eval_runtime": 144.9695,
|
| 824 |
+
"eval_samples_per_second": 34.49,
|
| 825 |
+
"eval_steps_per_second": 2.159,
|
| 826 |
+
"step": 8500
|
| 827 |
+
},
|
| 828 |
+
{
|
| 829 |
+
"epoch": 2.752,
|
| 830 |
+
"grad_norm": 0.1662980318069458,
|
| 831 |
+
"learning_rate": 9.372683688311512e-07,
|
| 832 |
+
"loss": 0.0089,
|
| 833 |
+
"step": 8600
|
| 834 |
+
},
|
| 835 |
+
{
|
| 836 |
+
"epoch": 2.784,
|
| 837 |
+
"grad_norm": 0.3024541735649109,
|
| 838 |
+
"learning_rate": 7.1234946812129e-07,
|
| 839 |
+
"loss": 0.0096,
|
| 840 |
+
"step": 8700
|
| 841 |
+
},
|
| 842 |
+
{
|
| 843 |
+
"epoch": 2.816,
|
| 844 |
+
"grad_norm": 0.39870765805244446,
|
| 845 |
+
"learning_rate": 5.178606532859453e-07,
|
| 846 |
+
"loss": 0.0094,
|
| 847 |
+
"step": 8800
|
| 848 |
+
},
|
| 849 |
+
{
|
| 850 |
+
"epoch": 2.848,
|
| 851 |
+
"grad_norm": 0.2740815281867981,
|
| 852 |
+
"learning_rate": 3.540456000700909e-07,
|
| 853 |
+
"loss": 0.0103,
|
| 854 |
+
"step": 8900
|
| 855 |
+
},
|
| 856 |
+
{
|
| 857 |
+
"epoch": 2.88,
|
| 858 |
+
"grad_norm": 0.47213834524154663,
|
| 859 |
+
"learning_rate": 2.211095529509738e-07,
|
| 860 |
+
"loss": 0.0089,
|
| 861 |
+
"step": 9000
|
| 862 |
+
},
|
| 863 |
+
{
|
| 864 |
+
"epoch": 2.88,
|
| 865 |
+
"eval_gen_len": 20.9772,
|
| 866 |
+
"eval_loss": 0.009878098033368587,
|
| 867 |
+
"eval_rouge1": 0.5132,
|
| 868 |
+
"eval_rouge2": 0.4508,
|
| 869 |
+
"eval_rougeL": 0.5121,
|
| 870 |
+
"eval_rougeLsum": 0.512,
|
| 871 |
+
"eval_runtime": 143.3601,
|
| 872 |
+
"eval_samples_per_second": 34.877,
|
| 873 |
+
"eval_steps_per_second": 2.183,
|
| 874 |
+
"step": 9000
|
| 875 |
+
}
|
| 876 |
+
],
|
| 877 |
+
"logging_steps": 100,
|
| 878 |
+
"max_steps": 9375,
|
| 879 |
+
"num_input_tokens_seen": 0,
|
| 880 |
+
"num_train_epochs": 3,
|
| 881 |
+
"save_steps": 500,
|
| 882 |
+
"stateful_callbacks": {
|
| 883 |
+
"EarlyStoppingCallback": {
|
| 884 |
+
"args": {
|
| 885 |
+
"early_stopping_patience": 3,
|
| 886 |
+
"early_stopping_threshold": 0.0
|
| 887 |
+
},
|
| 888 |
+
"attributes": {
|
| 889 |
+
"early_stopping_patience_counter": 0
|
| 890 |
+
}
|
| 891 |
+
},
|
| 892 |
+
"TrainerControl": {
|
| 893 |
+
"args": {
|
| 894 |
+
"should_epoch_stop": false,
|
| 895 |
+
"should_evaluate": false,
|
| 896 |
+
"should_log": false,
|
| 897 |
+
"should_save": true,
|
| 898 |
+
"should_training_stop": false
|
| 899 |
+
},
|
| 900 |
+
"attributes": {}
|
| 901 |
+
}
|
| 902 |
+
},
|
| 903 |
+
"total_flos": 1.154860858884096e+16,
|
| 904 |
+
"train_batch_size": 16,
|
| 905 |
+
"trial_name": null,
|
| 906 |
+
"trial_params": null
|
| 907 |
+
}
|
checkpoint-9000/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2d0804b816c1ade31e9b0e7d027931a2ac27eabce0b89f3a3acc0efcc369af28
|
| 3 |
+
size 5496
|
checkpoint-9000/vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-9375/config.json
ADDED
|
@@ -0,0 +1,73 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"activation_dropout": 0.1,
|
| 3 |
+
"activation_function": "gelu",
|
| 4 |
+
"add_bias_logits": false,
|
| 5 |
+
"add_final_layer_norm": false,
|
| 6 |
+
"architectures": [
|
| 7 |
+
"BartForConditionalGeneration"
|
| 8 |
+
],
|
| 9 |
+
"attention_dropout": 0.1,
|
| 10 |
+
"bos_token_id": 0,
|
| 11 |
+
"classif_dropout": 0.1,
|
| 12 |
+
"classifier_dropout": 0.0,
|
| 13 |
+
"d_model": 768,
|
| 14 |
+
"decoder_attention_heads": 12,
|
| 15 |
+
"decoder_ffn_dim": 3072,
|
| 16 |
+
"decoder_layerdrop": 0.0,
|
| 17 |
+
"decoder_layers": 6,
|
| 18 |
+
"decoder_start_token_id": 2,
|
| 19 |
+
"dropout": 0.1,
|
| 20 |
+
"early_stopping": null,
|
| 21 |
+
"encoder_attention_heads": 12,
|
| 22 |
+
"encoder_ffn_dim": 3072,
|
| 23 |
+
"encoder_layerdrop": 0.0,
|
| 24 |
+
"encoder_layers": 6,
|
| 25 |
+
"eos_token_id": 2,
|
| 26 |
+
"forced_eos_token_id": 2,
|
| 27 |
+
"gradient_checkpointing": false,
|
| 28 |
+
"id2label": {
|
| 29 |
+
"0": "LABEL_0",
|
| 30 |
+
"1": "LABEL_1",
|
| 31 |
+
"2": "LABEL_2"
|
| 32 |
+
},
|
| 33 |
+
"init_std": 0.02,
|
| 34 |
+
"is_encoder_decoder": true,
|
| 35 |
+
"label2id": {
|
| 36 |
+
"LABEL_0": 0,
|
| 37 |
+
"LABEL_1": 1,
|
| 38 |
+
"LABEL_2": 2
|
| 39 |
+
},
|
| 40 |
+
"max_position_embeddings": 1024,
|
| 41 |
+
"model_type": "bart",
|
| 42 |
+
"no_repeat_ngram_size": null,
|
| 43 |
+
"normalize_before": false,
|
| 44 |
+
"normalize_embedding": true,
|
| 45 |
+
"num_beams": null,
|
| 46 |
+
"num_hidden_layers": 6,
|
| 47 |
+
"pad_token_id": 1,
|
| 48 |
+
"scale_embedding": false,
|
| 49 |
+
"task_specific_params": {
|
| 50 |
+
"summarization": {
|
| 51 |
+
"length_penalty": 1.0,
|
| 52 |
+
"max_length": 128,
|
| 53 |
+
"min_length": 12,
|
| 54 |
+
"num_beams": 4
|
| 55 |
+
},
|
| 56 |
+
"summarization_cnn": {
|
| 57 |
+
"length_penalty": 2.0,
|
| 58 |
+
"max_length": 142,
|
| 59 |
+
"min_length": 56,
|
| 60 |
+
"num_beams": 4
|
| 61 |
+
},
|
| 62 |
+
"summarization_xsum": {
|
| 63 |
+
"length_penalty": 1.0,
|
| 64 |
+
"max_length": 62,
|
| 65 |
+
"min_length": 11,
|
| 66 |
+
"num_beams": 6
|
| 67 |
+
}
|
| 68 |
+
},
|
| 69 |
+
"torch_dtype": "float32",
|
| 70 |
+
"transformers_version": "4.55.4",
|
| 71 |
+
"use_cache": true,
|
| 72 |
+
"vocab_size": 50265
|
| 73 |
+
}
|
checkpoint-9375/generation_config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 0,
|
| 3 |
+
"decoder_start_token_id": 2,
|
| 4 |
+
"early_stopping": true,
|
| 5 |
+
"eos_token_id": 2,
|
| 6 |
+
"forced_bos_token_id": 0,
|
| 7 |
+
"forced_eos_token_id": 2,
|
| 8 |
+
"no_repeat_ngram_size": 3,
|
| 9 |
+
"num_beams": 4,
|
| 10 |
+
"pad_token_id": 1,
|
| 11 |
+
"transformers_version": "4.55.4"
|
| 12 |
+
}
|
checkpoint-9375/merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-9375/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0539d7a5539831a9708a040455c29ca68e77f0e1b2343e1f681048be23a27cf6
|
| 3 |
+
size 557912620
|
checkpoint-9375/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:72fe2cb142abf3b04511f698e9b79bdc833e8241d17f719a7a9e9bf7f8001b5b
|
| 3 |
+
size 1115579898
|
checkpoint-9375/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e7e58598e552f1485da103cd80bc03736232b1ecff16f76d98ee38ee655bd119
|
| 3 |
+
size 14244
|
checkpoint-9375/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:157a1e342bc4add90e936c7908881d49f6b07b28531474b18aec1d4e5b64aa59
|
| 3 |
+
size 1064
|
checkpoint-9375/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": true,
|
| 9 |
+
"rstrip": false,
|
| 10 |
+
"single_word": false
|
| 11 |
+
},
|
| 12 |
+
"pad_token": "<pad>",
|
| 13 |
+
"sep_token": "</s>",
|
| 14 |
+
"unk_token": "<unk>"
|
| 15 |
+
}
|
checkpoint-9375/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-9375/tokenizer_config.json
ADDED
|
@@ -0,0 +1,58 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"added_tokens_decoder": {
|
| 4 |
+
"0": {
|
| 5 |
+
"content": "<s>",
|
| 6 |
+
"lstrip": false,
|
| 7 |
+
"normalized": true,
|
| 8 |
+
"rstrip": false,
|
| 9 |
+
"single_word": false,
|
| 10 |
+
"special": true
|
| 11 |
+
},
|
| 12 |
+
"1": {
|
| 13 |
+
"content": "<pad>",
|
| 14 |
+
"lstrip": false,
|
| 15 |
+
"normalized": true,
|
| 16 |
+
"rstrip": false,
|
| 17 |
+
"single_word": false,
|
| 18 |
+
"special": true
|
| 19 |
+
},
|
| 20 |
+
"2": {
|
| 21 |
+
"content": "</s>",
|
| 22 |
+
"lstrip": false,
|
| 23 |
+
"normalized": true,
|
| 24 |
+
"rstrip": false,
|
| 25 |
+
"single_word": false,
|
| 26 |
+
"special": true
|
| 27 |
+
},
|
| 28 |
+
"3": {
|
| 29 |
+
"content": "<unk>",
|
| 30 |
+
"lstrip": false,
|
| 31 |
+
"normalized": true,
|
| 32 |
+
"rstrip": false,
|
| 33 |
+
"single_word": false,
|
| 34 |
+
"special": true
|
| 35 |
+
},
|
| 36 |
+
"50264": {
|
| 37 |
+
"content": "<mask>",
|
| 38 |
+
"lstrip": true,
|
| 39 |
+
"normalized": true,
|
| 40 |
+
"rstrip": false,
|
| 41 |
+
"single_word": false,
|
| 42 |
+
"special": true
|
| 43 |
+
}
|
| 44 |
+
},
|
| 45 |
+
"bos_token": "<s>",
|
| 46 |
+
"clean_up_tokenization_spaces": false,
|
| 47 |
+
"cls_token": "<s>",
|
| 48 |
+
"eos_token": "</s>",
|
| 49 |
+
"errors": "replace",
|
| 50 |
+
"extra_special_tokens": {},
|
| 51 |
+
"mask_token": "<mask>",
|
| 52 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 53 |
+
"pad_token": "<pad>",
|
| 54 |
+
"sep_token": "</s>",
|
| 55 |
+
"tokenizer_class": "BartTokenizer",
|
| 56 |
+
"trim_offsets": true,
|
| 57 |
+
"unk_token": "<unk>"
|
| 58 |
+
}
|
checkpoint-9375/trainer_state.json
ADDED
|
@@ -0,0 +1,928 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 9000,
|
| 3 |
+
"best_metric": 0.009878098033368587,
|
| 4 |
+
"best_model_checkpoint": "bart-base-finetune/checkpoint-9000",
|
| 5 |
+
"epoch": 3.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 9375,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.032,
|
| 14 |
+
"grad_norm": 3.6718413829803467,
|
| 15 |
+
"learning_rate": 9.900000000000002e-06,
|
| 16 |
+
"loss": 2.4853,
|
| 17 |
+
"step": 100
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.064,
|
| 21 |
+
"grad_norm": 2.339207649230957,
|
| 22 |
+
"learning_rate": 1.9900000000000003e-05,
|
| 23 |
+
"loss": 0.5438,
|
| 24 |
+
"step": 200
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.096,
|
| 28 |
+
"grad_norm": 1.8246095180511475,
|
| 29 |
+
"learning_rate": 2.9900000000000002e-05,
|
| 30 |
+
"loss": 0.3038,
|
| 31 |
+
"step": 300
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.128,
|
| 35 |
+
"grad_norm": 1.9325376749038696,
|
| 36 |
+
"learning_rate": 3.99e-05,
|
| 37 |
+
"loss": 0.2137,
|
| 38 |
+
"step": 400
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.16,
|
| 42 |
+
"grad_norm": 1.487189531326294,
|
| 43 |
+
"learning_rate": 4.99e-05,
|
| 44 |
+
"loss": 0.1689,
|
| 45 |
+
"step": 500
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.16,
|
| 49 |
+
"eval_gen_len": 20.9688,
|
| 50 |
+
"eval_loss": 0.10288307070732117,
|
| 51 |
+
"eval_rouge1": 0.4964,
|
| 52 |
+
"eval_rouge2": 0.4123,
|
| 53 |
+
"eval_rougeL": 0.4927,
|
| 54 |
+
"eval_rougeLsum": 0.4927,
|
| 55 |
+
"eval_runtime": 144.5729,
|
| 56 |
+
"eval_samples_per_second": 34.585,
|
| 57 |
+
"eval_steps_per_second": 2.165,
|
| 58 |
+
"step": 500
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.192,
|
| 62 |
+
"grad_norm": 0.9393866062164307,
|
| 63 |
+
"learning_rate": 4.998465033232545e-05,
|
| 64 |
+
"loss": 0.1361,
|
| 65 |
+
"step": 600
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.224,
|
| 69 |
+
"grad_norm": 1.4453977346420288,
|
| 70 |
+
"learning_rate": 4.9937998871479073e-05,
|
| 71 |
+
"loss": 0.1156,
|
| 72 |
+
"step": 700
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.256,
|
| 76 |
+
"grad_norm": 1.1997668743133545,
|
| 77 |
+
"learning_rate": 4.9860102500953126e-05,
|
| 78 |
+
"loss": 0.0988,
|
| 79 |
+
"step": 800
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.288,
|
| 83 |
+
"grad_norm": 1.3937125205993652,
|
| 84 |
+
"learning_rate": 4.975105881739415e-05,
|
| 85 |
+
"loss": 0.0889,
|
| 86 |
+
"step": 900
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.32,
|
| 90 |
+
"grad_norm": 1.3387874364852905,
|
| 91 |
+
"learning_rate": 4.961100444203071e-05,
|
| 92 |
+
"loss": 0.0793,
|
| 93 |
+
"step": 1000
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.32,
|
| 97 |
+
"eval_gen_len": 20.9698,
|
| 98 |
+
"eval_loss": 0.05177076905965805,
|
| 99 |
+
"eval_rouge1": 0.4988,
|
| 100 |
+
"eval_rouge2": 0.4239,
|
| 101 |
+
"eval_rougeL": 0.494,
|
| 102 |
+
"eval_rougeLsum": 0.494,
|
| 103 |
+
"eval_runtime": 144.6374,
|
| 104 |
+
"eval_samples_per_second": 34.569,
|
| 105 |
+
"eval_steps_per_second": 2.164,
|
| 106 |
+
"step": 1000
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 0.352,
|
| 110 |
+
"grad_norm": 1.226305365562439,
|
| 111 |
+
"learning_rate": 4.944011484950018e-05,
|
| 112 |
+
"loss": 0.0726,
|
| 113 |
+
"step": 1100
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 0.384,
|
| 117 |
+
"grad_norm": 1.3799694776535034,
|
| 118 |
+
"learning_rate": 4.9238604147995884e-05,
|
| 119 |
+
"loss": 0.0708,
|
| 120 |
+
"step": 1200
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 0.416,
|
| 124 |
+
"grad_norm": 1.0327191352844238,
|
| 125 |
+
"learning_rate": 4.900672481101019e-05,
|
| 126 |
+
"loss": 0.0675,
|
| 127 |
+
"step": 1300
|
| 128 |
+
},
|
| 129 |
+
{
|
| 130 |
+
"epoch": 0.448,
|
| 131 |
+
"grad_norm": 0.8962351083755493,
|
| 132 |
+
"learning_rate": 4.8744767361009594e-05,
|
| 133 |
+
"loss": 0.0604,
|
| 134 |
+
"step": 1400
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 0.48,
|
| 138 |
+
"grad_norm": 0.6496387720108032,
|
| 139 |
+
"learning_rate": 4.8453060005437934e-05,
|
| 140 |
+
"loss": 0.0568,
|
| 141 |
+
"step": 1500
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 0.48,
|
| 145 |
+
"eval_gen_len": 20.9742,
|
| 146 |
+
"eval_loss": 0.03353753313422203,
|
| 147 |
+
"eval_rouge1": 0.5059,
|
| 148 |
+
"eval_rouge2": 0.4356,
|
| 149 |
+
"eval_rougeL": 0.5032,
|
| 150 |
+
"eval_rougeLsum": 0.503,
|
| 151 |
+
"eval_runtime": 145.5239,
|
| 152 |
+
"eval_samples_per_second": 34.359,
|
| 153 |
+
"eval_steps_per_second": 2.151,
|
| 154 |
+
"step": 1500
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 0.512,
|
| 158 |
+
"grad_norm": 1.6739161014556885,
|
| 159 |
+
"learning_rate": 4.8131968225504245e-05,
|
| 160 |
+
"loss": 0.0539,
|
| 161 |
+
"step": 1600
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 0.544,
|
| 165 |
+
"grad_norm": 1.063624382019043,
|
| 166 |
+
"learning_rate": 4.778189431826991e-05,
|
| 167 |
+
"loss": 0.0502,
|
| 168 |
+
"step": 1700
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 0.576,
|
| 172 |
+
"grad_norm": 1.1451120376586914,
|
| 173 |
+
"learning_rate": 4.7403276892609224e-05,
|
| 174 |
+
"loss": 0.0477,
|
| 175 |
+
"step": 1800
|
| 176 |
+
},
|
| 177 |
+
{
|
| 178 |
+
"epoch": 0.608,
|
| 179 |
+
"grad_norm": 0.7193688750267029,
|
| 180 |
+
"learning_rate": 4.699659031967479e-05,
|
| 181 |
+
"loss": 0.0457,
|
| 182 |
+
"step": 1900
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 0.64,
|
| 186 |
+
"grad_norm": 0.9038898944854736,
|
| 187 |
+
"learning_rate": 4.656234413855613e-05,
|
| 188 |
+
"loss": 0.0434,
|
| 189 |
+
"step": 2000
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 0.64,
|
| 193 |
+
"eval_gen_len": 20.9714,
|
| 194 |
+
"eval_loss": 0.029211629182100296,
|
| 195 |
+
"eval_rouge1": 0.5076,
|
| 196 |
+
"eval_rouge2": 0.4388,
|
| 197 |
+
"eval_rougeL": 0.5055,
|
| 198 |
+
"eval_rougeLsum": 0.5054,
|
| 199 |
+
"eval_runtime": 145.669,
|
| 200 |
+
"eval_samples_per_second": 34.324,
|
| 201 |
+
"eval_steps_per_second": 2.149,
|
| 202 |
+
"step": 2000
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 0.672,
|
| 206 |
+
"grad_norm": 1.7222179174423218,
|
| 207 |
+
"learning_rate": 4.610108241787629e-05,
|
| 208 |
+
"loss": 0.0461,
|
| 209 |
+
"step": 2100
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 0.704,
|
| 213 |
+
"grad_norm": 0.6831910014152527,
|
| 214 |
+
"learning_rate": 4.561338307412629e-05,
|
| 215 |
+
"loss": 0.0438,
|
| 216 |
+
"step": 2200
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 0.736,
|
| 220 |
+
"grad_norm": 0.4437102675437927,
|
| 221 |
+
"learning_rate": 4.509985714759151e-05,
|
| 222 |
+
"loss": 0.0407,
|
| 223 |
+
"step": 2300
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"epoch": 0.768,
|
| 227 |
+
"grad_norm": 0.6408787965774536,
|
| 228 |
+
"learning_rate": 4.456114803677696e-05,
|
| 229 |
+
"loss": 0.0377,
|
| 230 |
+
"step": 2400
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 0.8,
|
| 234 |
+
"grad_norm": 0.7488101124763489,
|
| 235 |
+
"learning_rate": 4.3997930692291056e-05,
|
| 236 |
+
"loss": 0.0372,
|
| 237 |
+
"step": 2500
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 0.8,
|
| 241 |
+
"eval_gen_len": 20.9734,
|
| 242 |
+
"eval_loss": 0.02469678223133087,
|
| 243 |
+
"eval_rouge1": 0.5089,
|
| 244 |
+
"eval_rouge2": 0.4398,
|
| 245 |
+
"eval_rougeL": 0.5063,
|
| 246 |
+
"eval_rougeLsum": 0.5061,
|
| 247 |
+
"eval_runtime": 145.4374,
|
| 248 |
+
"eval_samples_per_second": 34.379,
|
| 249 |
+
"eval_steps_per_second": 2.152,
|
| 250 |
+
"step": 2500
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 0.832,
|
| 254 |
+
"grad_norm": 0.3353654742240906,
|
| 255 |
+
"learning_rate": 4.341091077119753e-05,
|
| 256 |
+
"loss": 0.0367,
|
| 257 |
+
"step": 2600
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 0.864,
|
| 261 |
+
"grad_norm": 0.853186845779419,
|
| 262 |
+
"learning_rate": 4.2800823752895055e-05,
|
| 263 |
+
"loss": 0.0345,
|
| 264 |
+
"step": 2700
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 0.896,
|
| 268 |
+
"grad_norm": 0.7370714545249939,
|
| 269 |
+
"learning_rate": 4.2168434017632516e-05,
|
| 270 |
+
"loss": 0.038,
|
| 271 |
+
"step": 2800
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 0.928,
|
| 275 |
+
"grad_norm": 0.5349611639976501,
|
| 276 |
+
"learning_rate": 4.151453388881413e-05,
|
| 277 |
+
"loss": 0.0321,
|
| 278 |
+
"step": 2900
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 0.96,
|
| 282 |
+
"grad_norm": 0.7140136957168579,
|
| 283 |
+
"learning_rate": 4.083994264029462e-05,
|
| 284 |
+
"loss": 0.0347,
|
| 285 |
+
"step": 3000
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 0.96,
|
| 289 |
+
"eval_gen_len": 20.9754,
|
| 290 |
+
"eval_loss": 0.01928527094423771,
|
| 291 |
+
"eval_rouge1": 0.5082,
|
| 292 |
+
"eval_rouge2": 0.4406,
|
| 293 |
+
"eval_rougeL": 0.5057,
|
| 294 |
+
"eval_rougeLsum": 0.5055,
|
| 295 |
+
"eval_runtime": 145.567,
|
| 296 |
+
"eval_samples_per_second": 34.348,
|
| 297 |
+
"eval_steps_per_second": 2.15,
|
| 298 |
+
"step": 3000
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"epoch": 0.992,
|
| 302 |
+
"grad_norm": 0.6110441088676453,
|
| 303 |
+
"learning_rate": 4.0145505469907984e-05,
|
| 304 |
+
"loss": 0.0325,
|
| 305 |
+
"step": 3100
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 1.024,
|
| 309 |
+
"grad_norm": 0.4845236539840698,
|
| 310 |
+
"learning_rate": 3.9432092440516075e-05,
|
| 311 |
+
"loss": 0.0276,
|
| 312 |
+
"step": 3200
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 1.056,
|
| 316 |
+
"grad_norm": 0.5893353819847107,
|
| 317 |
+
"learning_rate": 3.8700597389903664e-05,
|
| 318 |
+
"loss": 0.0262,
|
| 319 |
+
"step": 3300
|
| 320 |
+
},
|
| 321 |
+
{
|
| 322 |
+
"epoch": 1.088,
|
| 323 |
+
"grad_norm": 0.7486633658409119,
|
| 324 |
+
"learning_rate": 3.795193681088578e-05,
|
| 325 |
+
"loss": 0.0235,
|
| 326 |
+
"step": 3400
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 1.12,
|
| 330 |
+
"grad_norm": 0.886471688747406,
|
| 331 |
+
"learning_rate": 3.7187048703030597e-05,
|
| 332 |
+
"loss": 0.0235,
|
| 333 |
+
"step": 3500
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 1.12,
|
| 337 |
+
"eval_gen_len": 20.9758,
|
| 338 |
+
"eval_loss": 0.0190653745085001,
|
| 339 |
+
"eval_rouge1": 0.5087,
|
| 340 |
+
"eval_rouge2": 0.4431,
|
| 341 |
+
"eval_rougeL": 0.5072,
|
| 342 |
+
"eval_rougeLsum": 0.5071,
|
| 343 |
+
"eval_runtime": 145.2595,
|
| 344 |
+
"eval_samples_per_second": 34.421,
|
| 345 |
+
"eval_steps_per_second": 2.155,
|
| 346 |
+
"step": 3500
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 1.152,
|
| 350 |
+
"grad_norm": 1.1492247581481934,
|
| 351 |
+
"learning_rate": 3.640689139743628e-05,
|
| 352 |
+
"loss": 0.0234,
|
| 353 |
+
"step": 3600
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 1.184,
|
| 357 |
+
"grad_norm": 0.7707476019859314,
|
| 358 |
+
"learning_rate": 3.561244235603451e-05,
|
| 359 |
+
"loss": 0.0245,
|
| 360 |
+
"step": 3700
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 1.216,
|
| 364 |
+
"grad_norm": 0.8048288822174072,
|
| 365 |
+
"learning_rate": 3.4804696946924855e-05,
|
| 366 |
+
"loss": 0.0228,
|
| 367 |
+
"step": 3800
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 1.248,
|
| 371 |
+
"grad_norm": 0.7920108437538147,
|
| 372 |
+
"learning_rate": 3.398466719727445e-05,
|
| 373 |
+
"loss": 0.0227,
|
| 374 |
+
"step": 3900
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 1.28,
|
| 378 |
+
"grad_norm": 0.199005588889122,
|
| 379 |
+
"learning_rate": 3.3153380525345445e-05,
|
| 380 |
+
"loss": 0.0226,
|
| 381 |
+
"step": 4000
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 1.28,
|
| 385 |
+
"eval_gen_len": 20.975,
|
| 386 |
+
"eval_loss": 0.0172447320073843,
|
| 387 |
+
"eval_rouge1": 0.5091,
|
| 388 |
+
"eval_rouge2": 0.442,
|
| 389 |
+
"eval_rougeL": 0.5064,
|
| 390 |
+
"eval_rougeLsum": 0.5063,
|
| 391 |
+
"eval_runtime": 145.4831,
|
| 392 |
+
"eval_samples_per_second": 34.368,
|
| 393 |
+
"eval_steps_per_second": 2.151,
|
| 394 |
+
"step": 4000
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 1.312,
|
| 398 |
+
"grad_norm": 0.6734442114830017,
|
| 399 |
+
"learning_rate": 3.2311878453238966e-05,
|
| 400 |
+
"loss": 0.0207,
|
| 401 |
+
"step": 4100
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 1.3439999999999999,
|
| 405 |
+
"grad_norm": 0.5161114931106567,
|
| 406 |
+
"learning_rate": 3.146121530196824e-05,
|
| 407 |
+
"loss": 0.0216,
|
| 408 |
+
"step": 4200
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 1.376,
|
| 412 |
+
"grad_norm": 0.4449058175086975,
|
| 413 |
+
"learning_rate": 3.0602456870495973e-05,
|
| 414 |
+
"loss": 0.0197,
|
| 415 |
+
"step": 4300
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 1.408,
|
| 419 |
+
"grad_norm": 0.5729345083236694,
|
| 420 |
+
"learning_rate": 2.9736679100390947e-05,
|
| 421 |
+
"loss": 0.0208,
|
| 422 |
+
"step": 4400
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 1.44,
|
| 426 |
+
"grad_norm": 0.19021901488304138,
|
| 427 |
+
"learning_rate": 2.886496672777688e-05,
|
| 428 |
+
"loss": 0.0201,
|
| 429 |
+
"step": 4500
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 1.44,
|
| 433 |
+
"eval_gen_len": 20.9758,
|
| 434 |
+
"eval_loss": 0.014488083310425282,
|
| 435 |
+
"eval_rouge1": 0.5116,
|
| 436 |
+
"eval_rouge2": 0.4474,
|
| 437 |
+
"eval_rougeL": 0.5102,
|
| 438 |
+
"eval_rougeLsum": 0.5101,
|
| 439 |
+
"eval_runtime": 146.2423,
|
| 440 |
+
"eval_samples_per_second": 34.19,
|
| 441 |
+
"eval_steps_per_second": 2.14,
|
| 442 |
+
"step": 4500
|
| 443 |
+
},
|
| 444 |
+
{
|
| 445 |
+
"epoch": 1.472,
|
| 446 |
+
"grad_norm": 0.632882297039032,
|
| 447 |
+
"learning_rate": 2.7988411924262648e-05,
|
| 448 |
+
"loss": 0.0198,
|
| 449 |
+
"step": 4600
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 1.504,
|
| 453 |
+
"grad_norm": 0.37011951208114624,
|
| 454 |
+
"learning_rate": 2.71081129285564e-05,
|
| 455 |
+
"loss": 0.0205,
|
| 456 |
+
"step": 4700
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 1.536,
|
| 460 |
+
"grad_norm": 0.528691828250885,
|
| 461 |
+
"learning_rate": 2.622517267047835e-05,
|
| 462 |
+
"loss": 0.0204,
|
| 463 |
+
"step": 4800
|
| 464 |
+
},
|
| 465 |
+
{
|
| 466 |
+
"epoch": 1.568,
|
| 467 |
+
"grad_norm": 0.19730792939662933,
|
| 468 |
+
"learning_rate": 2.5340697389095973e-05,
|
| 469 |
+
"loss": 0.0193,
|
| 470 |
+
"step": 4900
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 1.6,
|
| 474 |
+
"grad_norm": 0.5517034530639648,
|
| 475 |
+
"learning_rate": 2.4455795246713044e-05,
|
| 476 |
+
"loss": 0.0164,
|
| 477 |
+
"step": 5000
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 1.6,
|
| 481 |
+
"eval_gen_len": 20.9766,
|
| 482 |
+
"eval_loss": 0.013988671824336052,
|
| 483 |
+
"eval_rouge1": 0.5114,
|
| 484 |
+
"eval_rouge2": 0.4467,
|
| 485 |
+
"eval_rougeL": 0.5099,
|
| 486 |
+
"eval_rougeLsum": 0.5098,
|
| 487 |
+
"eval_runtime": 145.249,
|
| 488 |
+
"eval_samples_per_second": 34.424,
|
| 489 |
+
"eval_steps_per_second": 2.155,
|
| 490 |
+
"step": 5000
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 1.6320000000000001,
|
| 494 |
+
"grad_norm": 0.6311265826225281,
|
| 495 |
+
"learning_rate": 2.3571574940449008e-05,
|
| 496 |
+
"loss": 0.0173,
|
| 497 |
+
"step": 5100
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 1.6640000000000001,
|
| 501 |
+
"grad_norm": 0.764806866645813,
|
| 502 |
+
"learning_rate": 2.2689144313148382e-05,
|
| 503 |
+
"loss": 0.0173,
|
| 504 |
+
"step": 5200
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 1.696,
|
| 508 |
+
"grad_norm": 1.1502550840377808,
|
| 509 |
+
"learning_rate": 2.180960896536036e-05,
|
| 510 |
+
"loss": 0.018,
|
| 511 |
+
"step": 5300
|
| 512 |
+
},
|
| 513 |
+
{
|
| 514 |
+
"epoch": 1.728,
|
| 515 |
+
"grad_norm": 0.21649418771266937,
|
| 516 |
+
"learning_rate": 2.0934070870127912e-05,
|
| 517 |
+
"loss": 0.0162,
|
| 518 |
+
"step": 5400
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 1.76,
|
| 522 |
+
"grad_norm": 0.3456017076969147,
|
| 523 |
+
"learning_rate": 2.0063626992321693e-05,
|
| 524 |
+
"loss": 0.017,
|
| 525 |
+
"step": 5500
|
| 526 |
+
},
|
| 527 |
+
{
|
| 528 |
+
"epoch": 1.76,
|
| 529 |
+
"eval_gen_len": 20.9772,
|
| 530 |
+
"eval_loss": 0.012542162090539932,
|
| 531 |
+
"eval_rouge1": 0.5119,
|
| 532 |
+
"eval_rouge2": 0.4484,
|
| 533 |
+
"eval_rougeL": 0.5108,
|
| 534 |
+
"eval_rougeLsum": 0.5106,
|
| 535 |
+
"eval_runtime": 144.5364,
|
| 536 |
+
"eval_samples_per_second": 34.593,
|
| 537 |
+
"eval_steps_per_second": 2.166,
|
| 538 |
+
"step": 5500
|
| 539 |
+
},
|
| 540 |
+
{
|
| 541 |
+
"epoch": 1.792,
|
| 542 |
+
"grad_norm": 0.5721284747123718,
|
| 543 |
+
"learning_rate": 1.919936791424879e-05,
|
| 544 |
+
"loss": 0.0176,
|
| 545 |
+
"step": 5600
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 1.8239999999999998,
|
| 549 |
+
"grad_norm": 0.26220762729644775,
|
| 550 |
+
"learning_rate": 1.834237646925815e-05,
|
| 551 |
+
"loss": 0.0152,
|
| 552 |
+
"step": 5700
|
| 553 |
+
},
|
| 554 |
+
{
|
| 555 |
+
"epoch": 1.8559999999999999,
|
| 556 |
+
"grad_norm": 0.271634966135025,
|
| 557 |
+
"learning_rate": 1.7493726385054743e-05,
|
| 558 |
+
"loss": 0.0152,
|
| 559 |
+
"step": 5800
|
| 560 |
+
},
|
| 561 |
+
{
|
| 562 |
+
"epoch": 1.888,
|
| 563 |
+
"grad_norm": 0.3826049566268921,
|
| 564 |
+
"learning_rate": 1.6654480938422208e-05,
|
| 565 |
+
"loss": 0.0151,
|
| 566 |
+
"step": 5900
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 1.92,
|
| 570 |
+
"grad_norm": 0.49469685554504395,
|
| 571 |
+
"learning_rate": 1.58256916230395e-05,
|
| 572 |
+
"loss": 0.0146,
|
| 573 |
+
"step": 6000
|
| 574 |
+
},
|
| 575 |
+
{
|
| 576 |
+
"epoch": 1.92,
|
| 577 |
+
"eval_gen_len": 20.976,
|
| 578 |
+
"eval_loss": 0.011809424497187138,
|
| 579 |
+
"eval_rouge1": 0.5115,
|
| 580 |
+
"eval_rouge2": 0.4481,
|
| 581 |
+
"eval_rougeL": 0.5103,
|
| 582 |
+
"eval_rougeLsum": 0.5101,
|
| 583 |
+
"eval_runtime": 144.9109,
|
| 584 |
+
"eval_samples_per_second": 34.504,
|
| 585 |
+
"eval_steps_per_second": 2.16,
|
| 586 |
+
"step": 6000
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"epoch": 1.952,
|
| 590 |
+
"grad_norm": 0.6852744221687317,
|
| 591 |
+
"learning_rate": 1.5008396832060606e-05,
|
| 592 |
+
"loss": 0.0161,
|
| 593 |
+
"step": 6100
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 1.984,
|
| 597 |
+
"grad_norm": 0.27392491698265076,
|
| 598 |
+
"learning_rate": 1.4203620557108022e-05,
|
| 599 |
+
"loss": 0.0146,
|
| 600 |
+
"step": 6200
|
| 601 |
+
},
|
| 602 |
+
{
|
| 603 |
+
"epoch": 2.016,
|
| 604 |
+
"grad_norm": 0.6940794587135315,
|
| 605 |
+
"learning_rate": 1.3412371105309856e-05,
|
| 606 |
+
"loss": 0.0132,
|
| 607 |
+
"step": 6300
|
| 608 |
+
},
|
| 609 |
+
{
|
| 610 |
+
"epoch": 2.048,
|
| 611 |
+
"grad_norm": 0.2785722613334656,
|
| 612 |
+
"learning_rate": 1.2635639835988195e-05,
|
| 613 |
+
"loss": 0.0106,
|
| 614 |
+
"step": 6400
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 2.08,
|
| 618 |
+
"grad_norm": 0.1548278033733368,
|
| 619 |
+
"learning_rate": 1.1874399918581388e-05,
|
| 620 |
+
"loss": 0.0114,
|
| 621 |
+
"step": 6500
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"epoch": 2.08,
|
| 625 |
+
"eval_gen_len": 20.9772,
|
| 626 |
+
"eval_loss": 0.010936998762190342,
|
| 627 |
+
"eval_rouge1": 0.5117,
|
| 628 |
+
"eval_rouge2": 0.4484,
|
| 629 |
+
"eval_rougeL": 0.5105,
|
| 630 |
+
"eval_rougeLsum": 0.5105,
|
| 631 |
+
"eval_runtime": 145.1325,
|
| 632 |
+
"eval_samples_per_second": 34.451,
|
| 633 |
+
"eval_steps_per_second": 2.157,
|
| 634 |
+
"step": 6500
|
| 635 |
+
},
|
| 636 |
+
{
|
| 637 |
+
"epoch": 2.112,
|
| 638 |
+
"grad_norm": 0.14601846039295197,
|
| 639 |
+
"learning_rate": 1.1129605113356508e-05,
|
| 640 |
+
"loss": 0.0111,
|
| 641 |
+
"step": 6600
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 2.144,
|
| 645 |
+
"grad_norm": 0.18163363635540009,
|
| 646 |
+
"learning_rate": 1.040218857643962e-05,
|
| 647 |
+
"loss": 0.0111,
|
| 648 |
+
"step": 6700
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 2.176,
|
| 652 |
+
"grad_norm": 0.2548365294933319,
|
| 653 |
+
"learning_rate": 9.693061690661118e-06,
|
| 654 |
+
"loss": 0.011,
|
| 655 |
+
"step": 6800
|
| 656 |
+
},
|
| 657 |
+
{
|
| 658 |
+
"epoch": 2.208,
|
| 659 |
+
"grad_norm": 0.21669094264507294,
|
| 660 |
+
"learning_rate": 9.00311292368079e-06,
|
| 661 |
+
"loss": 0.0111,
|
| 662 |
+
"step": 6900
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 2.24,
|
| 666 |
+
"grad_norm": 0.47503963112831116,
|
| 667 |
+
"learning_rate": 8.333206714823513e-06,
|
| 668 |
+
"loss": 0.0104,
|
| 669 |
+
"step": 7000
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 2.24,
|
| 673 |
+
"eval_gen_len": 20.9772,
|
| 674 |
+
"eval_loss": 0.01037654373794794,
|
| 675 |
+
"eval_rouge1": 0.5129,
|
| 676 |
+
"eval_rouge2": 0.4501,
|
| 677 |
+
"eval_rougeL": 0.5118,
|
| 678 |
+
"eval_rougeLsum": 0.5117,
|
| 679 |
+
"eval_runtime": 143.5577,
|
| 680 |
+
"eval_samples_per_second": 34.829,
|
| 681 |
+
"eval_steps_per_second": 2.18,
|
| 682 |
+
"step": 7000
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 2.2720000000000002,
|
| 686 |
+
"grad_norm": 0.7779979705810547,
|
| 687 |
+
"learning_rate": 7.684182392020015e-06,
|
| 688 |
+
"loss": 0.0109,
|
| 689 |
+
"step": 7100
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 2.304,
|
| 693 |
+
"grad_norm": 0.7133929133415222,
|
| 694 |
+
"learning_rate": 7.056853120209888e-06,
|
| 695 |
+
"loss": 0.0117,
|
| 696 |
+
"step": 7200
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 2.336,
|
| 700 |
+
"grad_norm": 0.5449277758598328,
|
| 701 |
+
"learning_rate": 6.45200488252426e-06,
|
| 702 |
+
"loss": 0.0097,
|
| 703 |
+
"step": 7300
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 2.368,
|
| 707 |
+
"grad_norm": 0.5456063747406006,
|
| 708 |
+
"learning_rate": 5.870395495524664e-06,
|
| 709 |
+
"loss": 0.0095,
|
| 710 |
+
"step": 7400
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 2.4,
|
| 714 |
+
"grad_norm": 0.21037012338638306,
|
| 715 |
+
"learning_rate": 5.312753659731909e-06,
|
| 716 |
+
"loss": 0.01,
|
| 717 |
+
"step": 7500
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 2.4,
|
| 721 |
+
"eval_gen_len": 20.9772,
|
| 722 |
+
"eval_loss": 0.010514349676668644,
|
| 723 |
+
"eval_rouge1": 0.5124,
|
| 724 |
+
"eval_rouge2": 0.4493,
|
| 725 |
+
"eval_rougeL": 0.5111,
|
| 726 |
+
"eval_rougeLsum": 0.5111,
|
| 727 |
+
"eval_runtime": 143.501,
|
| 728 |
+
"eval_samples_per_second": 34.843,
|
| 729 |
+
"eval_steps_per_second": 2.181,
|
| 730 |
+
"step": 7500
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 2.432,
|
| 734 |
+
"grad_norm": 0.5376931428909302,
|
| 735 |
+
"learning_rate": 4.779778046634537e-06,
|
| 736 |
+
"loss": 0.0114,
|
| 737 |
+
"step": 7600
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 2.464,
|
| 741 |
+
"grad_norm": 0.655579149723053,
|
| 742 |
+
"learning_rate": 4.272136423320786e-06,
|
| 743 |
+
"loss": 0.0099,
|
| 744 |
+
"step": 7700
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 2.496,
|
| 748 |
+
"grad_norm": 0.24749213457107544,
|
| 749 |
+
"learning_rate": 3.7904648158307465e-06,
|
| 750 |
+
"loss": 0.0088,
|
| 751 |
+
"step": 7800
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 2.528,
|
| 755 |
+
"grad_norm": 0.09235558658838272,
|
| 756 |
+
"learning_rate": 3.335366712277019e-06,
|
| 757 |
+
"loss": 0.0092,
|
| 758 |
+
"step": 7900
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 2.56,
|
| 762 |
+
"grad_norm": 0.4215800166130066,
|
| 763 |
+
"learning_rate": 2.907412306732288e-06,
|
| 764 |
+
"loss": 0.01,
|
| 765 |
+
"step": 8000
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 2.56,
|
| 769 |
+
"eval_gen_len": 20.9772,
|
| 770 |
+
"eval_loss": 0.009960692375898361,
|
| 771 |
+
"eval_rouge1": 0.5131,
|
| 772 |
+
"eval_rouge2": 0.4506,
|
| 773 |
+
"eval_rougeL": 0.5119,
|
| 774 |
+
"eval_rougeLsum": 0.5119,
|
| 775 |
+
"eval_runtime": 143.2943,
|
| 776 |
+
"eval_samples_per_second": 34.893,
|
| 777 |
+
"eval_steps_per_second": 2.184,
|
| 778 |
+
"step": 8000
|
| 779 |
+
},
|
| 780 |
+
{
|
| 781 |
+
"epoch": 2.592,
|
| 782 |
+
"grad_norm": 0.15820010006427765,
|
| 783 |
+
"learning_rate": 2.5071377848310595e-06,
|
| 784 |
+
"loss": 0.0092,
|
| 785 |
+
"step": 8100
|
| 786 |
+
},
|
| 787 |
+
{
|
| 788 |
+
"epoch": 2.624,
|
| 789 |
+
"grad_norm": 0.24991130828857422,
|
| 790 |
+
"learning_rate": 2.135044651980733e-06,
|
| 791 |
+
"loss": 0.0089,
|
| 792 |
+
"step": 8200
|
| 793 |
+
},
|
| 794 |
+
{
|
| 795 |
+
"epoch": 2.656,
|
| 796 |
+
"grad_norm": 0.5366423726081848,
|
| 797 |
+
"learning_rate": 1.7915991050236502e-06,
|
| 798 |
+
"loss": 0.0099,
|
| 799 |
+
"step": 8300
|
| 800 |
+
},
|
| 801 |
+
{
|
| 802 |
+
"epoch": 2.6879999999999997,
|
| 803 |
+
"grad_norm": 0.2854973375797272,
|
| 804 |
+
"learning_rate": 1.4772314481373838e-06,
|
| 805 |
+
"loss": 0.0096,
|
| 806 |
+
"step": 8400
|
| 807 |
+
},
|
| 808 |
+
{
|
| 809 |
+
"epoch": 2.7199999999999998,
|
| 810 |
+
"grad_norm": 0.24889616668224335,
|
| 811 |
+
"learning_rate": 1.1923355537050458e-06,
|
| 812 |
+
"loss": 0.0084,
|
| 813 |
+
"step": 8500
|
| 814 |
+
},
|
| 815 |
+
{
|
| 816 |
+
"epoch": 2.7199999999999998,
|
| 817 |
+
"eval_gen_len": 20.9772,
|
| 818 |
+
"eval_loss": 0.009925846941769123,
|
| 819 |
+
"eval_rouge1": 0.5132,
|
| 820 |
+
"eval_rouge2": 0.4505,
|
| 821 |
+
"eval_rougeL": 0.512,
|
| 822 |
+
"eval_rougeLsum": 0.5119,
|
| 823 |
+
"eval_runtime": 144.9695,
|
| 824 |
+
"eval_samples_per_second": 34.49,
|
| 825 |
+
"eval_steps_per_second": 2.159,
|
| 826 |
+
"step": 8500
|
| 827 |
+
},
|
| 828 |
+
{
|
| 829 |
+
"epoch": 2.752,
|
| 830 |
+
"grad_norm": 0.1662980318069458,
|
| 831 |
+
"learning_rate": 9.372683688311512e-07,
|
| 832 |
+
"loss": 0.0089,
|
| 833 |
+
"step": 8600
|
| 834 |
+
},
|
| 835 |
+
{
|
| 836 |
+
"epoch": 2.784,
|
| 837 |
+
"grad_norm": 0.3024541735649109,
|
| 838 |
+
"learning_rate": 7.1234946812129e-07,
|
| 839 |
+
"loss": 0.0096,
|
| 840 |
+
"step": 8700
|
| 841 |
+
},
|
| 842 |
+
{
|
| 843 |
+
"epoch": 2.816,
|
| 844 |
+
"grad_norm": 0.39870765805244446,
|
| 845 |
+
"learning_rate": 5.178606532859453e-07,
|
| 846 |
+
"loss": 0.0094,
|
| 847 |
+
"step": 8800
|
| 848 |
+
},
|
| 849 |
+
{
|
| 850 |
+
"epoch": 2.848,
|
| 851 |
+
"grad_norm": 0.2740815281867981,
|
| 852 |
+
"learning_rate": 3.540456000700909e-07,
|
| 853 |
+
"loss": 0.0103,
|
| 854 |
+
"step": 8900
|
| 855 |
+
},
|
| 856 |
+
{
|
| 857 |
+
"epoch": 2.88,
|
| 858 |
+
"grad_norm": 0.47213834524154663,
|
| 859 |
+
"learning_rate": 2.211095529509738e-07,
|
| 860 |
+
"loss": 0.0089,
|
| 861 |
+
"step": 9000
|
| 862 |
+
},
|
| 863 |
+
{
|
| 864 |
+
"epoch": 2.88,
|
| 865 |
+
"eval_gen_len": 20.9772,
|
| 866 |
+
"eval_loss": 0.009878098033368587,
|
| 867 |
+
"eval_rouge1": 0.5132,
|
| 868 |
+
"eval_rouge2": 0.4508,
|
| 869 |
+
"eval_rougeL": 0.5121,
|
| 870 |
+
"eval_rougeLsum": 0.512,
|
| 871 |
+
"eval_runtime": 143.3601,
|
| 872 |
+
"eval_samples_per_second": 34.877,
|
| 873 |
+
"eval_steps_per_second": 2.183,
|
| 874 |
+
"step": 9000
|
| 875 |
+
},
|
| 876 |
+
{
|
| 877 |
+
"epoch": 2.912,
|
| 878 |
+
"grad_norm": 0.4264777898788452,
|
| 879 |
+
"learning_rate": 1.1921906798654513e-07,
|
| 880 |
+
"loss": 0.0098,
|
| 881 |
+
"step": 9100
|
| 882 |
+
},
|
| 883 |
+
{
|
| 884 |
+
"epoch": 2.944,
|
| 885 |
+
"grad_norm": 0.2921827435493469,
|
| 886 |
+
"learning_rate": 4.850180413679406e-08,
|
| 887 |
+
"loss": 0.0087,
|
| 888 |
+
"step": 9200
|
| 889 |
+
},
|
| 890 |
+
{
|
| 891 |
+
"epoch": 2.976,
|
| 892 |
+
"grad_norm": 0.5423845648765564,
|
| 893 |
+
"learning_rate": 9.046363319339945e-09,
|
| 894 |
+
"loss": 0.0096,
|
| 895 |
+
"step": 9300
|
| 896 |
+
}
|
| 897 |
+
],
|
| 898 |
+
"logging_steps": 100,
|
| 899 |
+
"max_steps": 9375,
|
| 900 |
+
"num_input_tokens_seen": 0,
|
| 901 |
+
"num_train_epochs": 3,
|
| 902 |
+
"save_steps": 500,
|
| 903 |
+
"stateful_callbacks": {
|
| 904 |
+
"EarlyStoppingCallback": {
|
| 905 |
+
"args": {
|
| 906 |
+
"early_stopping_patience": 3,
|
| 907 |
+
"early_stopping_threshold": 0.0
|
| 908 |
+
},
|
| 909 |
+
"attributes": {
|
| 910 |
+
"early_stopping_patience_counter": 0
|
| 911 |
+
}
|
| 912 |
+
},
|
| 913 |
+
"TrainerControl": {
|
| 914 |
+
"args": {
|
| 915 |
+
"should_epoch_stop": false,
|
| 916 |
+
"should_evaluate": false,
|
| 917 |
+
"should_log": false,
|
| 918 |
+
"should_save": true,
|
| 919 |
+
"should_training_stop": true
|
| 920 |
+
},
|
| 921 |
+
"attributes": {}
|
| 922 |
+
}
|
| 923 |
+
},
|
| 924 |
+
"total_flos": 1.20259083829248e+16,
|
| 925 |
+
"train_batch_size": 16,
|
| 926 |
+
"trial_name": null,
|
| 927 |
+
"trial_params": null
|
| 928 |
+
}
|
checkpoint-9375/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2d0804b816c1ade31e9b0e7d027931a2ac27eabce0b89f3a3acc0efcc369af28
|
| 3 |
+
size 5496
|
checkpoint-9375/vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
config.json
ADDED
|
@@ -0,0 +1,73 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"activation_dropout": 0.1,
|
| 3 |
+
"activation_function": "gelu",
|
| 4 |
+
"add_bias_logits": false,
|
| 5 |
+
"add_final_layer_norm": false,
|
| 6 |
+
"architectures": [
|
| 7 |
+
"BartForConditionalGeneration"
|
| 8 |
+
],
|
| 9 |
+
"attention_dropout": 0.1,
|
| 10 |
+
"bos_token_id": 0,
|
| 11 |
+
"classif_dropout": 0.1,
|
| 12 |
+
"classifier_dropout": 0.0,
|
| 13 |
+
"d_model": 768,
|
| 14 |
+
"decoder_attention_heads": 12,
|
| 15 |
+
"decoder_ffn_dim": 3072,
|
| 16 |
+
"decoder_layerdrop": 0.0,
|
| 17 |
+
"decoder_layers": 6,
|
| 18 |
+
"decoder_start_token_id": 2,
|
| 19 |
+
"dropout": 0.1,
|
| 20 |
+
"early_stopping": null,
|
| 21 |
+
"encoder_attention_heads": 12,
|
| 22 |
+
"encoder_ffn_dim": 3072,
|
| 23 |
+
"encoder_layerdrop": 0.0,
|
| 24 |
+
"encoder_layers": 6,
|
| 25 |
+
"eos_token_id": 2,
|
| 26 |
+
"forced_eos_token_id": 2,
|
| 27 |
+
"gradient_checkpointing": false,
|
| 28 |
+
"id2label": {
|
| 29 |
+
"0": "LABEL_0",
|
| 30 |
+
"1": "LABEL_1",
|
| 31 |
+
"2": "LABEL_2"
|
| 32 |
+
},
|
| 33 |
+
"init_std": 0.02,
|
| 34 |
+
"is_encoder_decoder": true,
|
| 35 |
+
"label2id": {
|
| 36 |
+
"LABEL_0": 0,
|
| 37 |
+
"LABEL_1": 1,
|
| 38 |
+
"LABEL_2": 2
|
| 39 |
+
},
|
| 40 |
+
"max_position_embeddings": 1024,
|
| 41 |
+
"model_type": "bart",
|
| 42 |
+
"no_repeat_ngram_size": null,
|
| 43 |
+
"normalize_before": false,
|
| 44 |
+
"normalize_embedding": true,
|
| 45 |
+
"num_beams": null,
|
| 46 |
+
"num_hidden_layers": 6,
|
| 47 |
+
"pad_token_id": 1,
|
| 48 |
+
"scale_embedding": false,
|
| 49 |
+
"task_specific_params": {
|
| 50 |
+
"summarization": {
|
| 51 |
+
"length_penalty": 1.0,
|
| 52 |
+
"max_length": 128,
|
| 53 |
+
"min_length": 12,
|
| 54 |
+
"num_beams": 4
|
| 55 |
+
},
|
| 56 |
+
"summarization_cnn": {
|
| 57 |
+
"length_penalty": 2.0,
|
| 58 |
+
"max_length": 142,
|
| 59 |
+
"min_length": 56,
|
| 60 |
+
"num_beams": 4
|
| 61 |
+
},
|
| 62 |
+
"summarization_xsum": {
|
| 63 |
+
"length_penalty": 1.0,
|
| 64 |
+
"max_length": 62,
|
| 65 |
+
"min_length": 11,
|
| 66 |
+
"num_beams": 6
|
| 67 |
+
}
|
| 68 |
+
},
|
| 69 |
+
"torch_dtype": "float32",
|
| 70 |
+
"transformers_version": "4.55.4",
|
| 71 |
+
"use_cache": true,
|
| 72 |
+
"vocab_size": 50265
|
| 73 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 0,
|
| 3 |
+
"decoder_start_token_id": 2,
|
| 4 |
+
"early_stopping": true,
|
| 5 |
+
"eos_token_id": 2,
|
| 6 |
+
"forced_bos_token_id": 0,
|
| 7 |
+
"forced_eos_token_id": 2,
|
| 8 |
+
"no_repeat_ngram_size": 3,
|
| 9 |
+
"num_beams": 4,
|
| 10 |
+
"pad_token_id": 1,
|
| 11 |
+
"transformers_version": "4.55.4"
|
| 12 |
+
}
|
merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2aec4c08609cfd6d0c21bab230dba30501b0c4d262386f1cc13e06a064d4dbac
|
| 3 |
+
size 557912620
|
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": true,
|
| 9 |
+
"rstrip": false,
|
| 10 |
+
"single_word": false
|
| 11 |
+
},
|
| 12 |
+
"pad_token": "<pad>",
|
| 13 |
+
"sep_token": "</s>",
|
| 14 |
+
"unk_token": "<unk>"
|
| 15 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,58 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"added_tokens_decoder": {
|
| 4 |
+
"0": {
|
| 5 |
+
"content": "<s>",
|
| 6 |
+
"lstrip": false,
|
| 7 |
+
"normalized": true,
|
| 8 |
+
"rstrip": false,
|
| 9 |
+
"single_word": false,
|
| 10 |
+
"special": true
|
| 11 |
+
},
|
| 12 |
+
"1": {
|
| 13 |
+
"content": "<pad>",
|
| 14 |
+
"lstrip": false,
|
| 15 |
+
"normalized": true,
|
| 16 |
+
"rstrip": false,
|
| 17 |
+
"single_word": false,
|
| 18 |
+
"special": true
|
| 19 |
+
},
|
| 20 |
+
"2": {
|
| 21 |
+
"content": "</s>",
|
| 22 |
+
"lstrip": false,
|
| 23 |
+
"normalized": true,
|
| 24 |
+
"rstrip": false,
|
| 25 |
+
"single_word": false,
|
| 26 |
+
"special": true
|
| 27 |
+
},
|
| 28 |
+
"3": {
|
| 29 |
+
"content": "<unk>",
|
| 30 |
+
"lstrip": false,
|
| 31 |
+
"normalized": true,
|
| 32 |
+
"rstrip": false,
|
| 33 |
+
"single_word": false,
|
| 34 |
+
"special": true
|
| 35 |
+
},
|
| 36 |
+
"50264": {
|
| 37 |
+
"content": "<mask>",
|
| 38 |
+
"lstrip": true,
|
| 39 |
+
"normalized": true,
|
| 40 |
+
"rstrip": false,
|
| 41 |
+
"single_word": false,
|
| 42 |
+
"special": true
|
| 43 |
+
}
|
| 44 |
+
},
|
| 45 |
+
"bos_token": "<s>",
|
| 46 |
+
"clean_up_tokenization_spaces": false,
|
| 47 |
+
"cls_token": "<s>",
|
| 48 |
+
"eos_token": "</s>",
|
| 49 |
+
"errors": "replace",
|
| 50 |
+
"extra_special_tokens": {},
|
| 51 |
+
"mask_token": "<mask>",
|
| 52 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 53 |
+
"pad_token": "<pad>",
|
| 54 |
+
"sep_token": "</s>",
|
| 55 |
+
"tokenizer_class": "BartTokenizer",
|
| 56 |
+
"trim_offsets": true,
|
| 57 |
+
"unk_token": "<unk>"
|
| 58 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2d0804b816c1ade31e9b0e7d027931a2ac27eabce0b89f3a3acc0efcc369af28
|
| 3 |
+
size 5496
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|