upload checkpoint files
Browse files- config.json +70 -0
- generation_config.json +16 -0
- merges.txt +0 -0
- pytorch_model.bin +3 -0
- special_tokens_map.json +15 -0
- tokenizer.json +0 -0
- tokenizer_config.json +15 -0
- trainer_state.json +436 -0
- vocab.json +0 -0
config.json
ADDED
|
@@ -0,0 +1,70 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "yuyang/bart-large-cnn-finetuned-newsroom",
|
| 3 |
+
"_num_labels": 3,
|
| 4 |
+
"activation_dropout": 0.0,
|
| 5 |
+
"activation_function": "gelu",
|
| 6 |
+
"add_final_layer_norm": false,
|
| 7 |
+
"architectures": [
|
| 8 |
+
"NFBart"
|
| 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 |
+
"early_stopping": true,
|
| 22 |
+
"encoder_attention_heads": 16,
|
| 23 |
+
"encoder_ffn_dim": 4096,
|
| 24 |
+
"encoder_layerdrop": 0.0,
|
| 25 |
+
"encoder_layers": 12,
|
| 26 |
+
"eos_token_id": 2,
|
| 27 |
+
"force_bos_token_to_be_generated": true,
|
| 28 |
+
"forced_bos_token_id": 0,
|
| 29 |
+
"forced_eos_token_id": 2,
|
| 30 |
+
"gradient_checkpointing": false,
|
| 31 |
+
"id2label": {
|
| 32 |
+
"0": "LABEL_0",
|
| 33 |
+
"1": "LABEL_1",
|
| 34 |
+
"2": "LABEL_2"
|
| 35 |
+
},
|
| 36 |
+
"init_std": 0.02,
|
| 37 |
+
"is_encoder_decoder": true,
|
| 38 |
+
"label2id": {
|
| 39 |
+
"LABEL_0": 0,
|
| 40 |
+
"LABEL_1": 1,
|
| 41 |
+
"LABEL_2": 2
|
| 42 |
+
},
|
| 43 |
+
"length_penalty": 2.0,
|
| 44 |
+
"max_length": 142,
|
| 45 |
+
"max_position_embeddings": 1024,
|
| 46 |
+
"min_length": 56,
|
| 47 |
+
"model_type": "bart",
|
| 48 |
+
"no_repeat_ngram_size": 3,
|
| 49 |
+
"normalize_before": false,
|
| 50 |
+
"num_beams": 4,
|
| 51 |
+
"num_hidden_layers": 12,
|
| 52 |
+
"output_past": true,
|
| 53 |
+
"pad_token_id": 1,
|
| 54 |
+
"prefix": " ",
|
| 55 |
+
"scale_embedding": false,
|
| 56 |
+
"task_specific_params": {
|
| 57 |
+
"summarization": {
|
| 58 |
+
"early_stopping": true,
|
| 59 |
+
"length_penalty": 2.0,
|
| 60 |
+
"max_length": 142,
|
| 61 |
+
"min_length": 56,
|
| 62 |
+
"no_repeat_ngram_size": 3,
|
| 63 |
+
"num_beams": 4
|
| 64 |
+
}
|
| 65 |
+
},
|
| 66 |
+
"torch_dtype": "float32",
|
| 67 |
+
"transformers_version": "4.27.1",
|
| 68 |
+
"use_cache": true,
|
| 69 |
+
"vocab_size": 50265
|
| 70 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 0,
|
| 4 |
+
"decoder_start_token_id": 2,
|
| 5 |
+
"early_stopping": true,
|
| 6 |
+
"eos_token_id": 2,
|
| 7 |
+
"forced_bos_token_id": 0,
|
| 8 |
+
"forced_eos_token_id": 2,
|
| 9 |
+
"length_penalty": 2.0,
|
| 10 |
+
"max_length": 142,
|
| 11 |
+
"min_length": 56,
|
| 12 |
+
"no_repeat_ngram_size": 3,
|
| 13 |
+
"num_beams": 4,
|
| 14 |
+
"pad_token_id": 1,
|
| 15 |
+
"transformers_version": "4.27.1"
|
| 16 |
+
}
|
merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2dc78b5a8775d3396e1aa6e5fb384135b8eef8afbc44f42f1d82aea3f7f0c543
|
| 3 |
+
size 1660501409
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": "<s>",
|
| 3 |
+
"cls_token": "<s>",
|
| 4 |
+
"eos_token": "</s>",
|
| 5 |
+
"mask_token": {
|
| 6 |
+
"content": "<mask>",
|
| 7 |
+
"lstrip": true,
|
| 8 |
+
"normalized": false,
|
| 9 |
+
"rstrip": false,
|
| 10 |
+
"single_word": false
|
| 11 |
+
},
|
| 12 |
+
"pad_token": "<pad>",
|
| 13 |
+
"sep_token": "</s>",
|
| 14 |
+
"unk_token": "<unk>"
|
| 15 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"bos_token": "<s>",
|
| 4 |
+
"cls_token": "<s>",
|
| 5 |
+
"eos_token": "</s>",
|
| 6 |
+
"errors": "replace",
|
| 7 |
+
"mask_token": "<mask>",
|
| 8 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 9 |
+
"pad_token": "<pad>",
|
| 10 |
+
"sep_token": "</s>",
|
| 11 |
+
"special_tokens_map_file": "/home/shenx/yang6367/.cache/huggingface/hub/models--yuyang--bart-large-cnn-finetuned-newsroom/snapshots/5684014ca8aaf3e5e5604206981dca7a86f6f265/special_tokens_map.json",
|
| 12 |
+
"tokenizer_class": "BartTokenizer",
|
| 13 |
+
"trim_offsets": true,
|
| 14 |
+
"unk_token": "<unk>"
|
| 15 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,436 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 5.6603827476501465,
|
| 3 |
+
"best_model_checkpoint": "../checkpoints/nf-bart-newsroom-rqnsf/checkpoint-60000",
|
| 4 |
+
"epoch": 0.9648008490247472,
|
| 5 |
+
"global_step": 60000,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.03,
|
| 12 |
+
"gate_score": 0.371,
|
| 13 |
+
"learning_rate": 4.9725e-05,
|
| 14 |
+
"loss": 592.8312,
|
| 15 |
+
"nf_loss": 591.7519,
|
| 16 |
+
"ppl": 3.3519,
|
| 17 |
+
"step": 2000
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.06,
|
| 21 |
+
"gate_score": 0.2007,
|
| 22 |
+
"learning_rate": 4.834853544667631e-05,
|
| 23 |
+
"loss": 333.2925,
|
| 24 |
+
"nf_loss": 332.3035,
|
| 25 |
+
"ppl": 2.8873,
|
| 26 |
+
"step": 4000
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.06,
|
| 30 |
+
"eval_loss": 252.84817504882812,
|
| 31 |
+
"eval_nf_loss": 251.2677459716797,
|
| 32 |
+
"eval_perplexity": 6.363341331481934,
|
| 33 |
+
"eval_runtime": 4381.8372,
|
| 34 |
+
"eval_samples_per_second": 24.837,
|
| 35 |
+
"eval_steps_per_second": 1.552,
|
| 36 |
+
"step": 4000
|
| 37 |
+
},
|
| 38 |
+
{
|
| 39 |
+
"epoch": 0.1,
|
| 40 |
+
"gate_score": 0.1135,
|
| 41 |
+
"learning_rate": 4.6687102294439184e-05,
|
| 42 |
+
"loss": 211.0105,
|
| 43 |
+
"nf_loss": 209.9559,
|
| 44 |
+
"ppl": 3.1158,
|
| 45 |
+
"step": 6000
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.13,
|
| 49 |
+
"gate_score": 0.0666,
|
| 50 |
+
"learning_rate": 4.5025669142202066e-05,
|
| 51 |
+
"loss": 134.1864,
|
| 52 |
+
"nf_loss": 133.0995,
|
| 53 |
+
"ppl": 3.2307,
|
| 54 |
+
"step": 8000
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 0.13,
|
| 58 |
+
"eval_loss": 102.85702514648438,
|
| 59 |
+
"eval_nf_loss": 101.2651596069336,
|
| 60 |
+
"eval_perplexity": 6.448662757873535,
|
| 61 |
+
"eval_runtime": 4005.1516,
|
| 62 |
+
"eval_samples_per_second": 27.173,
|
| 63 |
+
"eval_steps_per_second": 1.698,
|
| 64 |
+
"step": 8000
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 0.16,
|
| 68 |
+
"gate_score": 0.0352,
|
| 69 |
+
"learning_rate": 4.3365897423117185e-05,
|
| 70 |
+
"loss": 95.272,
|
| 71 |
+
"nf_loss": 94.1643,
|
| 72 |
+
"ppl": 3.2978,
|
| 73 |
+
"step": 10000
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.19,
|
| 77 |
+
"gate_score": 0.023,
|
| 78 |
+
"learning_rate": 4.170446427088006e-05,
|
| 79 |
+
"loss": 68.625,
|
| 80 |
+
"nf_loss": 67.5129,
|
| 81 |
+
"ppl": 3.324,
|
| 82 |
+
"step": 12000
|
| 83 |
+
},
|
| 84 |
+
{
|
| 85 |
+
"epoch": 0.19,
|
| 86 |
+
"eval_loss": 53.623291015625,
|
| 87 |
+
"eval_nf_loss": 52.02442169189453,
|
| 88 |
+
"eval_perplexity": 6.536867618560791,
|
| 89 |
+
"eval_runtime": 3924.6092,
|
| 90 |
+
"eval_samples_per_second": 27.731,
|
| 91 |
+
"eval_steps_per_second": 1.733,
|
| 92 |
+
"step": 12000
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.23,
|
| 96 |
+
"gate_score": 0.023,
|
| 97 |
+
"learning_rate": 4.004386183521906e-05,
|
| 98 |
+
"loss": 49.4308,
|
| 99 |
+
"nf_loss": 48.3272,
|
| 100 |
+
"ppl": 3.2964,
|
| 101 |
+
"step": 14000
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.26,
|
| 105 |
+
"gate_score": 0.0219,
|
| 106 |
+
"learning_rate": 3.8383259399558055e-05,
|
| 107 |
+
"loss": 36.663,
|
| 108 |
+
"nf_loss": 35.5419,
|
| 109 |
+
"ppl": 3.3952,
|
| 110 |
+
"step": 16000
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.26,
|
| 114 |
+
"eval_loss": 29.72612762451172,
|
| 115 |
+
"eval_nf_loss": 28.141469955444336,
|
| 116 |
+
"eval_perplexity": 6.444383144378662,
|
| 117 |
+
"eval_runtime": 3904.4886,
|
| 118 |
+
"eval_samples_per_second": 27.874,
|
| 119 |
+
"eval_steps_per_second": 1.742,
|
| 120 |
+
"step": 16000
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 0.29,
|
| 124 |
+
"gate_score": 0.0183,
|
| 125 |
+
"learning_rate": 3.672265696389706e-05,
|
| 126 |
+
"loss": 30.1161,
|
| 127 |
+
"nf_loss": 29.0084,
|
| 128 |
+
"ppl": 3.3204,
|
| 129 |
+
"step": 18000
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.32,
|
| 133 |
+
"gate_score": 0.0188,
|
| 134 |
+
"learning_rate": 3.5061223811659935e-05,
|
| 135 |
+
"loss": 25.2734,
|
| 136 |
+
"nf_loss": 24.1661,
|
| 137 |
+
"ppl": 3.3019,
|
| 138 |
+
"step": 20000
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 0.32,
|
| 142 |
+
"eval_loss": 21.10334014892578,
|
| 143 |
+
"eval_nf_loss": 19.523048400878906,
|
| 144 |
+
"eval_perplexity": 6.385104656219482,
|
| 145 |
+
"eval_runtime": 3951.5734,
|
| 146 |
+
"eval_samples_per_second": 27.541,
|
| 147 |
+
"eval_steps_per_second": 1.721,
|
| 148 |
+
"step": 20000
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 0.35,
|
| 152 |
+
"gate_score": 0.0189,
|
| 153 |
+
"learning_rate": 3.340062137599894e-05,
|
| 154 |
+
"loss": 21.8002,
|
| 155 |
+
"nf_loss": 20.6893,
|
| 156 |
+
"ppl": 3.3211,
|
| 157 |
+
"step": 22000
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.39,
|
| 161 |
+
"gate_score": 0.0186,
|
| 162 |
+
"learning_rate": 3.1740018940337936e-05,
|
| 163 |
+
"loss": 19.5679,
|
| 164 |
+
"nf_loss": 18.4664,
|
| 165 |
+
"ppl": 3.2967,
|
| 166 |
+
"step": 24000
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 0.39,
|
| 170 |
+
"eval_loss": 16.757369995117188,
|
| 171 |
+
"eval_nf_loss": 15.157673835754395,
|
| 172 |
+
"eval_perplexity": 6.50905179977417,
|
| 173 |
+
"eval_runtime": 4060.1325,
|
| 174 |
+
"eval_samples_per_second": 26.805,
|
| 175 |
+
"eval_steps_per_second": 1.675,
|
| 176 |
+
"step": 24000
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 0.42,
|
| 180 |
+
"gate_score": 0.0169,
|
| 181 |
+
"learning_rate": 3.0079416504676937e-05,
|
| 182 |
+
"loss": 17.5505,
|
| 183 |
+
"nf_loss": 16.4471,
|
| 184 |
+
"ppl": 3.2866,
|
| 185 |
+
"step": 26000
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.45,
|
| 189 |
+
"gate_score": 0.0169,
|
| 190 |
+
"learning_rate": 2.8417983352439813e-05,
|
| 191 |
+
"loss": 16.0243,
|
| 192 |
+
"nf_loss": 14.9265,
|
| 193 |
+
"ppl": 3.2771,
|
| 194 |
+
"step": 28000
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 0.45,
|
| 198 |
+
"eval_loss": 13.916472434997559,
|
| 199 |
+
"eval_nf_loss": 12.336326599121094,
|
| 200 |
+
"eval_perplexity": 6.333822250366211,
|
| 201 |
+
"eval_runtime": 4073.3885,
|
| 202 |
+
"eval_samples_per_second": 26.718,
|
| 203 |
+
"eval_steps_per_second": 1.67,
|
| 204 |
+
"step": 28000
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 0.48,
|
| 208 |
+
"gate_score": 0.0164,
|
| 209 |
+
"learning_rate": 2.6757380916778813e-05,
|
| 210 |
+
"loss": 14.7704,
|
| 211 |
+
"nf_loss": 13.6695,
|
| 212 |
+
"ppl": 3.2909,
|
| 213 |
+
"step": 30000
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.51,
|
| 217 |
+
"gate_score": 0.0165,
|
| 218 |
+
"learning_rate": 2.509677848111781e-05,
|
| 219 |
+
"loss": 13.8487,
|
| 220 |
+
"nf_loss": 12.7566,
|
| 221 |
+
"ppl": 3.2604,
|
| 222 |
+
"step": 32000
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 0.51,
|
| 226 |
+
"eval_loss": 12.310139656066895,
|
| 227 |
+
"eval_nf_loss": 10.705780029296875,
|
| 228 |
+
"eval_perplexity": 6.516045570373535,
|
| 229 |
+
"eval_runtime": 3950.6921,
|
| 230 |
+
"eval_samples_per_second": 27.548,
|
| 231 |
+
"eval_steps_per_second": 1.722,
|
| 232 |
+
"step": 32000
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 0.55,
|
| 236 |
+
"gate_score": 0.0163,
|
| 237 |
+
"learning_rate": 2.3435345328880693e-05,
|
| 238 |
+
"loss": 13.021,
|
| 239 |
+
"nf_loss": 11.939,
|
| 240 |
+
"ppl": 3.2217,
|
| 241 |
+
"step": 34000
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.58,
|
| 245 |
+
"gate_score": 0.0164,
|
| 246 |
+
"learning_rate": 2.1774742893219694e-05,
|
| 247 |
+
"loss": 12.3656,
|
| 248 |
+
"nf_loss": 11.2731,
|
| 249 |
+
"ppl": 3.2491,
|
| 250 |
+
"step": 36000
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 0.58,
|
| 254 |
+
"eval_loss": 11.076099395751953,
|
| 255 |
+
"eval_nf_loss": 9.518845558166504,
|
| 256 |
+
"eval_perplexity": 6.16482400894165,
|
| 257 |
+
"eval_runtime": 3933.1054,
|
| 258 |
+
"eval_samples_per_second": 27.671,
|
| 259 |
+
"eval_steps_per_second": 1.729,
|
| 260 |
+
"step": 36000
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 0.61,
|
| 264 |
+
"gate_score": 0.0169,
|
| 265 |
+
"learning_rate": 2.0113309740982573e-05,
|
| 266 |
+
"loss": 11.834,
|
| 267 |
+
"nf_loss": 10.7422,
|
| 268 |
+
"ppl": 3.248,
|
| 269 |
+
"step": 38000
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.64,
|
| 273 |
+
"gate_score": 0.017,
|
| 274 |
+
"learning_rate": 1.845270730532157e-05,
|
| 275 |
+
"loss": 11.301,
|
| 276 |
+
"nf_loss": 10.2211,
|
| 277 |
+
"ppl": 3.2176,
|
| 278 |
+
"step": 40000
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 0.64,
|
| 282 |
+
"eval_loss": 9.942972183227539,
|
| 283 |
+
"eval_nf_loss": 8.422438621520996,
|
| 284 |
+
"eval_perplexity": 5.888382911682129,
|
| 285 |
+
"eval_runtime": 3905.9353,
|
| 286 |
+
"eval_samples_per_second": 27.863,
|
| 287 |
+
"eval_steps_per_second": 1.741,
|
| 288 |
+
"step": 40000
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 0.68,
|
| 292 |
+
"gate_score": 0.0178,
|
| 293 |
+
"learning_rate": 1.679293558623669e-05,
|
| 294 |
+
"loss": 10.826,
|
| 295 |
+
"nf_loss": 9.7634,
|
| 296 |
+
"ppl": 3.1432,
|
| 297 |
+
"step": 42000
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.71,
|
| 301 |
+
"gate_score": 0.0167,
|
| 302 |
+
"learning_rate": 1.513150243399957e-05,
|
| 303 |
+
"loss": 10.5722,
|
| 304 |
+
"nf_loss": 9.4924,
|
| 305 |
+
"ppl": 3.2112,
|
| 306 |
+
"step": 44000
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 0.71,
|
| 310 |
+
"eval_loss": 9.34610366821289,
|
| 311 |
+
"eval_nf_loss": 7.819733619689941,
|
| 312 |
+
"eval_perplexity": 5.929094314575195,
|
| 313 |
+
"eval_runtime": 3809.9667,
|
| 314 |
+
"eval_samples_per_second": 28.565,
|
| 315 |
+
"eval_steps_per_second": 1.785,
|
| 316 |
+
"step": 44000
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 0.74,
|
| 320 |
+
"gate_score": 0.0164,
|
| 321 |
+
"learning_rate": 1.3470899998338569e-05,
|
| 322 |
+
"loss": 10.1982,
|
| 323 |
+
"nf_loss": 9.1308,
|
| 324 |
+
"ppl": 3.1538,
|
| 325 |
+
"step": 46000
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.77,
|
| 329 |
+
"gate_score": 0.0156,
|
| 330 |
+
"learning_rate": 1.1810297562677566e-05,
|
| 331 |
+
"loss": 9.9637,
|
| 332 |
+
"nf_loss": 8.9113,
|
| 333 |
+
"ppl": 3.1019,
|
| 334 |
+
"step": 48000
|
| 335 |
+
},
|
| 336 |
+
{
|
| 337 |
+
"epoch": 0.77,
|
| 338 |
+
"eval_loss": 9.272148132324219,
|
| 339 |
+
"eval_nf_loss": 7.770688056945801,
|
| 340 |
+
"eval_perplexity": 5.756178855895996,
|
| 341 |
+
"eval_runtime": 3799.2449,
|
| 342 |
+
"eval_samples_per_second": 28.646,
|
| 343 |
+
"eval_steps_per_second": 1.79,
|
| 344 |
+
"step": 48000
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 0.8,
|
| 348 |
+
"gate_score": 0.0159,
|
| 349 |
+
"learning_rate": 1.0148864410440447e-05,
|
| 350 |
+
"loss": 9.7627,
|
| 351 |
+
"nf_loss": 8.701,
|
| 352 |
+
"ppl": 3.1339,
|
| 353 |
+
"step": 50000
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.84,
|
| 357 |
+
"gate_score": 0.015,
|
| 358 |
+
"learning_rate": 8.488261974779444e-06,
|
| 359 |
+
"loss": 9.5228,
|
| 360 |
+
"nf_loss": 8.4605,
|
| 361 |
+
"ppl": 3.1399,
|
| 362 |
+
"step": 52000
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 0.84,
|
| 366 |
+
"eval_loss": 8.568826675415039,
|
| 367 |
+
"eval_nf_loss": 7.074533462524414,
|
| 368 |
+
"eval_perplexity": 5.70849609375,
|
| 369 |
+
"eval_runtime": 3805.5723,
|
| 370 |
+
"eval_samples_per_second": 28.598,
|
| 371 |
+
"eval_steps_per_second": 1.787,
|
| 372 |
+
"step": 52000
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 0.87,
|
| 376 |
+
"gate_score": 0.0143,
|
| 377 |
+
"learning_rate": 6.827659539118445e-06,
|
| 378 |
+
"loss": 9.3407,
|
| 379 |
+
"nf_loss": 8.2955,
|
| 380 |
+
"ppl": 3.0762,
|
| 381 |
+
"step": 54000
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.9,
|
| 385 |
+
"gate_score": 0.0148,
|
| 386 |
+
"learning_rate": 5.167057103457443e-06,
|
| 387 |
+
"loss": 9.1988,
|
| 388 |
+
"nf_loss": 8.1602,
|
| 389 |
+
"ppl": 3.058,
|
| 390 |
+
"step": 56000
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 0.9,
|
| 394 |
+
"eval_loss": 8.280665397644043,
|
| 395 |
+
"eval_nf_loss": 6.778278827667236,
|
| 396 |
+
"eval_perplexity": 5.754031658172607,
|
| 397 |
+
"eval_runtime": 3795.1994,
|
| 398 |
+
"eval_samples_per_second": 28.676,
|
| 399 |
+
"eval_steps_per_second": 1.792,
|
| 400 |
+
"step": 56000
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 0.93,
|
| 404 |
+
"gate_score": 0.0146,
|
| 405 |
+
"learning_rate": 3.5056239512203227e-06,
|
| 406 |
+
"loss": 9.0572,
|
| 407 |
+
"nf_loss": 8.0245,
|
| 408 |
+
"ppl": 3.0413,
|
| 409 |
+
"step": 58000
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.96,
|
| 413 |
+
"gate_score": 0.0138,
|
| 414 |
+
"learning_rate": 1.8450215155593216e-06,
|
| 415 |
+
"loss": 9.0079,
|
| 416 |
+
"nf_loss": 7.9878,
|
| 417 |
+
"ppl": 3.002,
|
| 418 |
+
"step": 60000
|
| 419 |
+
},
|
| 420 |
+
{
|
| 421 |
+
"epoch": 0.96,
|
| 422 |
+
"eval_loss": 8.222879409790039,
|
| 423 |
+
"eval_nf_loss": 6.735788822174072,
|
| 424 |
+
"eval_perplexity": 5.6603827476501465,
|
| 425 |
+
"eval_runtime": 3672.6524,
|
| 426 |
+
"eval_samples_per_second": 29.633,
|
| 427 |
+
"eval_steps_per_second": 1.852,
|
| 428 |
+
"step": 60000
|
| 429 |
+
}
|
| 430 |
+
],
|
| 431 |
+
"max_steps": 62189,
|
| 432 |
+
"num_train_epochs": 1,
|
| 433 |
+
"total_flos": 1.06597659967488e+18,
|
| 434 |
+
"trial_name": null,
|
| 435 |
+
"trial_params": null
|
| 436 |
+
}
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|