Upload folder using huggingface_hub
Browse files- added_tokens.json +0 -0
- config.json +75 -0
- generation_config.json +12 -0
- merges.txt +0 -0
- optimizer.pt +3 -0
- pytorch_model.bin +3 -0
- rng_state.pth +3 -0
- scaler.pt +3 -0
- scheduler.pt +3 -0
- special_tokens_map.json +15 -0
- tokenizer.json +0 -0
- tokenizer_config.json +15 -0
- trainer_state.json +1051 -0
- training_args.bin +3 -0
- vocab.json +0 -0
added_tokens.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
config.json
ADDED
|
@@ -0,0 +1,75 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "voidful/bart-base-unit",
|
| 3 |
+
"activation_dropout": 0.1,
|
| 4 |
+
"activation_function": "gelu",
|
| 5 |
+
"add_bias_logits": false,
|
| 6 |
+
"add_final_layer_norm": false,
|
| 7 |
+
"architectures": [
|
| 8 |
+
"BartForConditionalGeneration"
|
| 9 |
+
],
|
| 10 |
+
"attention_dropout": 0.1,
|
| 11 |
+
"bos_token_id": 0,
|
| 12 |
+
"classif_dropout": 0.1,
|
| 13 |
+
"classifier_dropout": 0.0,
|
| 14 |
+
"d_model": 768,
|
| 15 |
+
"decoder_attention_heads": 12,
|
| 16 |
+
"decoder_ffn_dim": 3072,
|
| 17 |
+
"decoder_layerdrop": 0.0,
|
| 18 |
+
"decoder_layers": 6,
|
| 19 |
+
"decoder_start_token_id": 2,
|
| 20 |
+
"dropout": 0.1,
|
| 21 |
+
"early_stopping": true,
|
| 22 |
+
"encoder_attention_heads": 12,
|
| 23 |
+
"encoder_ffn_dim": 3072,
|
| 24 |
+
"encoder_layerdrop": 0.0,
|
| 25 |
+
"encoder_layers": 6,
|
| 26 |
+
"eos_token_id": 2,
|
| 27 |
+
"forced_bos_token_id": 0,
|
| 28 |
+
"forced_eos_token_id": 2,
|
| 29 |
+
"gradient_checkpointing": false,
|
| 30 |
+
"id2label": {
|
| 31 |
+
"0": "LABEL_0",
|
| 32 |
+
"1": "LABEL_1",
|
| 33 |
+
"2": "LABEL_2"
|
| 34 |
+
},
|
| 35 |
+
"init_std": 0.02,
|
| 36 |
+
"is_encoder_decoder": true,
|
| 37 |
+
"label2id": {
|
| 38 |
+
"LABEL_0": 0,
|
| 39 |
+
"LABEL_1": 1,
|
| 40 |
+
"LABEL_2": 2
|
| 41 |
+
},
|
| 42 |
+
"max_position_embeddings": 1024,
|
| 43 |
+
"model_type": "bart",
|
| 44 |
+
"no_repeat_ngram_size": 3,
|
| 45 |
+
"normalize_before": false,
|
| 46 |
+
"normalize_embedding": true,
|
| 47 |
+
"num_beams": 4,
|
| 48 |
+
"num_hidden_layers": 6,
|
| 49 |
+
"pad_token_id": 1,
|
| 50 |
+
"scale_embedding": false,
|
| 51 |
+
"task_specific_params": {
|
| 52 |
+
"summarization": {
|
| 53 |
+
"length_penalty": 1.0,
|
| 54 |
+
"max_length": 128,
|
| 55 |
+
"min_length": 12,
|
| 56 |
+
"num_beams": 4
|
| 57 |
+
},
|
| 58 |
+
"summarization_cnn": {
|
| 59 |
+
"length_penalty": 2.0,
|
| 60 |
+
"max_length": 142,
|
| 61 |
+
"min_length": 56,
|
| 62 |
+
"num_beams": 4
|
| 63 |
+
},
|
| 64 |
+
"summarization_xsum": {
|
| 65 |
+
"length_penalty": 1.0,
|
| 66 |
+
"max_length": 62,
|
| 67 |
+
"min_length": 11,
|
| 68 |
+
"num_beams": 6
|
| 69 |
+
}
|
| 70 |
+
},
|
| 71 |
+
"torch_dtype": "float32",
|
| 72 |
+
"transformers_version": "4.27.3",
|
| 73 |
+
"use_cache": true,
|
| 74 |
+
"vocab_size": 59481
|
| 75 |
+
}
|
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": "4.27.3"
|
| 12 |
+
}
|
merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:09fa73e2c7b84135da0819cb3361ad29ed2a0d4133676d43e0e1f27fbe7eae32
|
| 3 |
+
size 1172138949
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3aae6b0de46d0a627b0d56586c1f74c2280b9728d454b35bcca69d1eb87dcfef
|
| 3 |
+
size 586319645
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8cdf02ff2b94ffef909b1acbcaa810beab6bba3d6fbcf672e6bd8d38024c7ad7
|
| 3 |
+
size 14575
|
scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6c216b88b94b8ee1ba153e4fb1108af5133b16dded5ecff8c628612c2cdcbaa0
|
| 3 |
+
size 557
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:57d33ee3ba84428f73b94c45c4b6650bef0e76d8131fe686577c80d9f8ac7c23
|
| 3 |
+
size 627
|
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": 1024,
|
| 9 |
+
"pad_token": "<pad>",
|
| 10 |
+
"sep_token": "</s>",
|
| 11 |
+
"special_tokens_map_file": null,
|
| 12 |
+
"tokenizer_class": "BartTokenizer",
|
| 13 |
+
"trim_offsets": true,
|
| 14 |
+
"unk_token": "<unk>"
|
| 15 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1051 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 9.784097580066533,
|
| 5 |
+
"global_step": 75000,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.07,
|
| 12 |
+
"learning_rate": 8.072407045009784e-06,
|
| 13 |
+
"loss": 6.7755,
|
| 14 |
+
"step": 500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.13,
|
| 18 |
+
"learning_rate": 1.622635355512068e-05,
|
| 19 |
+
"loss": 4.3924,
|
| 20 |
+
"step": 1000
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.2,
|
| 24 |
+
"learning_rate": 2.4380300065231574e-05,
|
| 25 |
+
"loss": 3.9807,
|
| 26 |
+
"step": 1500
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.26,
|
| 30 |
+
"learning_rate": 3.253424657534247e-05,
|
| 31 |
+
"loss": 3.7959,
|
| 32 |
+
"step": 2000
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.33,
|
| 36 |
+
"learning_rate": 4.068819308545336e-05,
|
| 37 |
+
"loss": 3.6875,
|
| 38 |
+
"step": 2500
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.39,
|
| 42 |
+
"learning_rate": 4.884213959556426e-05,
|
| 43 |
+
"loss": 3.6157,
|
| 44 |
+
"step": 3000
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.46,
|
| 48 |
+
"learning_rate": 5.699608610567515e-05,
|
| 49 |
+
"loss": 3.5719,
|
| 50 |
+
"step": 3500
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.52,
|
| 54 |
+
"learning_rate": 6.515003261578603e-05,
|
| 55 |
+
"loss": 3.5222,
|
| 56 |
+
"step": 4000
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.59,
|
| 60 |
+
"learning_rate": 7.330397912589694e-05,
|
| 61 |
+
"loss": 3.498,
|
| 62 |
+
"step": 4500
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.65,
|
| 66 |
+
"learning_rate": 8.145792563600784e-05,
|
| 67 |
+
"loss": 3.4647,
|
| 68 |
+
"step": 5000
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.65,
|
| 72 |
+
"eval_loss": 3.4406068325042725,
|
| 73 |
+
"eval_runtime": 5633.1588,
|
| 74 |
+
"eval_samples_per_second": 0.432,
|
| 75 |
+
"eval_steps_per_second": 0.054,
|
| 76 |
+
"eval_wer": 1.15852260825739,
|
| 77 |
+
"step": 5000
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 0.72,
|
| 81 |
+
"learning_rate": 8.961187214611873e-05,
|
| 82 |
+
"loss": 3.4349,
|
| 83 |
+
"step": 5500
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 0.78,
|
| 87 |
+
"learning_rate": 9.776581865622962e-05,
|
| 88 |
+
"loss": 3.4181,
|
| 89 |
+
"step": 6000
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 0.85,
|
| 93 |
+
"learning_rate": 9.948523781162257e-05,
|
| 94 |
+
"loss": 3.3879,
|
| 95 |
+
"step": 6500
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 0.91,
|
| 99 |
+
"learning_rate": 9.87761989846564e-05,
|
| 100 |
+
"loss": 3.3797,
|
| 101 |
+
"step": 7000
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.98,
|
| 105 |
+
"learning_rate": 9.806716015769025e-05,
|
| 106 |
+
"loss": 3.3632,
|
| 107 |
+
"step": 7500
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 1.04,
|
| 111 |
+
"learning_rate": 9.735812133072407e-05,
|
| 112 |
+
"loss": 3.3488,
|
| 113 |
+
"step": 8000
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 1.11,
|
| 117 |
+
"learning_rate": 9.664908250375791e-05,
|
| 118 |
+
"loss": 3.3378,
|
| 119 |
+
"step": 8500
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 1.17,
|
| 123 |
+
"learning_rate": 9.594004367679175e-05,
|
| 124 |
+
"loss": 3.3113,
|
| 125 |
+
"step": 9000
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 1.24,
|
| 129 |
+
"learning_rate": 9.523100484982558e-05,
|
| 130 |
+
"loss": 3.3031,
|
| 131 |
+
"step": 9500
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 1.3,
|
| 135 |
+
"learning_rate": 9.452338410051335e-05,
|
| 136 |
+
"loss": 3.2819,
|
| 137 |
+
"step": 10000
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 1.3,
|
| 141 |
+
"eval_loss": 3.261105537414551,
|
| 142 |
+
"eval_runtime": 7279.2824,
|
| 143 |
+
"eval_samples_per_second": 0.334,
|
| 144 |
+
"eval_steps_per_second": 0.042,
|
| 145 |
+
"eval_wer": 1.7922252072031033,
|
| 146 |
+
"step": 10000
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 1.37,
|
| 150 |
+
"learning_rate": 9.381434527354719e-05,
|
| 151 |
+
"loss": 3.2647,
|
| 152 |
+
"step": 10500
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 1.44,
|
| 156 |
+
"learning_rate": 9.310530644658103e-05,
|
| 157 |
+
"loss": 3.2264,
|
| 158 |
+
"step": 11000
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 1.5,
|
| 162 |
+
"learning_rate": 9.239626761961485e-05,
|
| 163 |
+
"loss": 3.2075,
|
| 164 |
+
"step": 11500
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 1.57,
|
| 168 |
+
"learning_rate": 9.168722879264869e-05,
|
| 169 |
+
"loss": 3.1744,
|
| 170 |
+
"step": 12000
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 1.63,
|
| 174 |
+
"learning_rate": 9.097818996568253e-05,
|
| 175 |
+
"loss": 3.1484,
|
| 176 |
+
"step": 12500
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 1.7,
|
| 180 |
+
"learning_rate": 9.026915113871636e-05,
|
| 181 |
+
"loss": 3.1259,
|
| 182 |
+
"step": 13000
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 1.76,
|
| 186 |
+
"learning_rate": 8.95601123117502e-05,
|
| 187 |
+
"loss": 3.1157,
|
| 188 |
+
"step": 13500
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 1.83,
|
| 192 |
+
"learning_rate": 8.885107348478403e-05,
|
| 193 |
+
"loss": 3.1022,
|
| 194 |
+
"step": 14000
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 1.89,
|
| 198 |
+
"learning_rate": 8.814203465781786e-05,
|
| 199 |
+
"loss": 3.0838,
|
| 200 |
+
"step": 14500
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 1.96,
|
| 204 |
+
"learning_rate": 8.74329958308517e-05,
|
| 205 |
+
"loss": 3.0741,
|
| 206 |
+
"step": 15000
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 1.96,
|
| 210 |
+
"eval_loss": 3.0150177478790283,
|
| 211 |
+
"eval_runtime": 7749.0478,
|
| 212 |
+
"eval_samples_per_second": 0.314,
|
| 213 |
+
"eval_steps_per_second": 0.039,
|
| 214 |
+
"eval_wer": 1.6811482450050355,
|
| 215 |
+
"step": 15000
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 2.02,
|
| 219 |
+
"learning_rate": 8.672395700388554e-05,
|
| 220 |
+
"loss": 3.0585,
|
| 221 |
+
"step": 15500
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 2.09,
|
| 225 |
+
"learning_rate": 8.601491817691936e-05,
|
| 226 |
+
"loss": 3.0499,
|
| 227 |
+
"step": 16000
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 2.15,
|
| 231 |
+
"learning_rate": 8.530729742760714e-05,
|
| 232 |
+
"loss": 3.0364,
|
| 233 |
+
"step": 16500
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 2.22,
|
| 237 |
+
"learning_rate": 8.459825860064097e-05,
|
| 238 |
+
"loss": 3.0311,
|
| 239 |
+
"step": 17000
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 2.28,
|
| 243 |
+
"learning_rate": 8.388921977367481e-05,
|
| 244 |
+
"loss": 3.0206,
|
| 245 |
+
"step": 17500
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 2.35,
|
| 249 |
+
"learning_rate": 8.318018094670864e-05,
|
| 250 |
+
"loss": 3.0151,
|
| 251 |
+
"step": 18000
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 2.41,
|
| 255 |
+
"learning_rate": 8.247114211974248e-05,
|
| 256 |
+
"loss": 3.015,
|
| 257 |
+
"step": 18500
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 2.48,
|
| 261 |
+
"learning_rate": 8.176352137043025e-05,
|
| 262 |
+
"loss": 3.0009,
|
| 263 |
+
"step": 19000
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 2.54,
|
| 267 |
+
"learning_rate": 8.105448254346409e-05,
|
| 268 |
+
"loss": 2.9991,
|
| 269 |
+
"step": 19500
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 2.61,
|
| 273 |
+
"learning_rate": 8.034544371649792e-05,
|
| 274 |
+
"loss": 2.9897,
|
| 275 |
+
"step": 20000
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 2.61,
|
| 279 |
+
"eval_loss": 2.9467122554779053,
|
| 280 |
+
"eval_runtime": 7497.8191,
|
| 281 |
+
"eval_samples_per_second": 0.324,
|
| 282 |
+
"eval_steps_per_second": 0.041,
|
| 283 |
+
"eval_wer": 1.880129997594331,
|
| 284 |
+
"step": 20000
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 2.67,
|
| 288 |
+
"learning_rate": 7.963640488953175e-05,
|
| 289 |
+
"loss": 2.9933,
|
| 290 |
+
"step": 20500
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 2.74,
|
| 294 |
+
"learning_rate": 7.89273660625656e-05,
|
| 295 |
+
"loss": 2.9868,
|
| 296 |
+
"step": 21000
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 2.8,
|
| 300 |
+
"learning_rate": 7.821832723559942e-05,
|
| 301 |
+
"loss": 2.9774,
|
| 302 |
+
"step": 21500
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 2.87,
|
| 306 |
+
"learning_rate": 7.750928840863326e-05,
|
| 307 |
+
"loss": 2.9887,
|
| 308 |
+
"step": 22000
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 2.94,
|
| 312 |
+
"learning_rate": 7.68002495816671e-05,
|
| 313 |
+
"loss": 2.9771,
|
| 314 |
+
"step": 22500
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 3.0,
|
| 318 |
+
"learning_rate": 7.609121075470092e-05,
|
| 319 |
+
"loss": 2.9693,
|
| 320 |
+
"step": 23000
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 3.07,
|
| 324 |
+
"learning_rate": 7.538500808304263e-05,
|
| 325 |
+
"loss": 2.9536,
|
| 326 |
+
"step": 23500
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 3.13,
|
| 330 |
+
"learning_rate": 7.467596925607647e-05,
|
| 331 |
+
"loss": 2.9497,
|
| 332 |
+
"step": 24000
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 3.2,
|
| 336 |
+
"learning_rate": 7.39669304291103e-05,
|
| 337 |
+
"loss": 2.9556,
|
| 338 |
+
"step": 24500
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 3.26,
|
| 342 |
+
"learning_rate": 7.325930967979807e-05,
|
| 343 |
+
"loss": 2.9472,
|
| 344 |
+
"step": 25000
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 3.26,
|
| 348 |
+
"eval_loss": 2.9073405265808105,
|
| 349 |
+
"eval_runtime": 7894.0702,
|
| 350 |
+
"eval_samples_per_second": 0.308,
|
| 351 |
+
"eval_steps_per_second": 0.039,
|
| 352 |
+
"eval_wer": 1.7444502680537224,
|
| 353 |
+
"step": 25000
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 3.33,
|
| 357 |
+
"learning_rate": 7.25502708528319e-05,
|
| 358 |
+
"loss": 2.9421,
|
| 359 |
+
"step": 25500
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 3.39,
|
| 363 |
+
"learning_rate": 7.184123202586573e-05,
|
| 364 |
+
"loss": 2.9412,
|
| 365 |
+
"step": 26000
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 3.46,
|
| 369 |
+
"learning_rate": 7.113219319889958e-05,
|
| 370 |
+
"loss": 2.9363,
|
| 371 |
+
"step": 26500
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 3.52,
|
| 375 |
+
"learning_rate": 7.042315437193341e-05,
|
| 376 |
+
"loss": 2.9404,
|
| 377 |
+
"step": 27000
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 3.59,
|
| 381 |
+
"learning_rate": 6.971411554496725e-05,
|
| 382 |
+
"loss": 2.9243,
|
| 383 |
+
"step": 27500
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 3.65,
|
| 387 |
+
"learning_rate": 6.900507671800109e-05,
|
| 388 |
+
"loss": 2.9242,
|
| 389 |
+
"step": 28000
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 3.72,
|
| 393 |
+
"learning_rate": 6.829603789103491e-05,
|
| 394 |
+
"loss": 2.9393,
|
| 395 |
+
"step": 28500
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 3.78,
|
| 399 |
+
"learning_rate": 6.758699906406875e-05,
|
| 400 |
+
"loss": 2.933,
|
| 401 |
+
"step": 29000
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 3.85,
|
| 405 |
+
"learning_rate": 6.687796023710259e-05,
|
| 406 |
+
"loss": 2.9206,
|
| 407 |
+
"step": 29500
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 3.91,
|
| 411 |
+
"learning_rate": 6.617033948779036e-05,
|
| 412 |
+
"loss": 2.9237,
|
| 413 |
+
"step": 30000
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 3.91,
|
| 417 |
+
"eval_loss": 2.881319999694824,
|
| 418 |
+
"eval_runtime": 8126.3615,
|
| 419 |
+
"eval_samples_per_second": 0.299,
|
| 420 |
+
"eval_steps_per_second": 0.037,
|
| 421 |
+
"eval_wer": 1.5856239783910557,
|
| 422 |
+
"step": 30000
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 3.98,
|
| 426 |
+
"learning_rate": 6.546130066082419e-05,
|
| 427 |
+
"loss": 2.9193,
|
| 428 |
+
"step": 30500
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 4.04,
|
| 432 |
+
"learning_rate": 6.475226183385803e-05,
|
| 433 |
+
"loss": 2.9006,
|
| 434 |
+
"step": 31000
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 4.11,
|
| 438 |
+
"learning_rate": 6.404322300689187e-05,
|
| 439 |
+
"loss": 2.9101,
|
| 440 |
+
"step": 31500
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 4.17,
|
| 444 |
+
"learning_rate": 6.333418417992569e-05,
|
| 445 |
+
"loss": 2.8993,
|
| 446 |
+
"step": 32000
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 4.24,
|
| 450 |
+
"learning_rate": 6.262514535295953e-05,
|
| 451 |
+
"loss": 2.9036,
|
| 452 |
+
"step": 32500
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 4.31,
|
| 456 |
+
"learning_rate": 6.191610652599337e-05,
|
| 457 |
+
"loss": 2.906,
|
| 458 |
+
"step": 33000
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 4.37,
|
| 462 |
+
"learning_rate": 6.12070676990272e-05,
|
| 463 |
+
"loss": 2.9025,
|
| 464 |
+
"step": 33500
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 4.44,
|
| 468 |
+
"learning_rate": 6.049802887206104e-05,
|
| 469 |
+
"loss": 2.9036,
|
| 470 |
+
"step": 34000
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 4.5,
|
| 474 |
+
"learning_rate": 5.9788990045094874e-05,
|
| 475 |
+
"loss": 2.8951,
|
| 476 |
+
"step": 34500
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 4.57,
|
| 480 |
+
"learning_rate": 5.907995121812871e-05,
|
| 481 |
+
"loss": 2.8923,
|
| 482 |
+
"step": 35000
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 4.57,
|
| 486 |
+
"eval_loss": 2.864828109741211,
|
| 487 |
+
"eval_runtime": 8189.637,
|
| 488 |
+
"eval_samples_per_second": 0.297,
|
| 489 |
+
"eval_steps_per_second": 0.037,
|
| 490 |
+
"eval_wer": 1.4885282519750724,
|
| 491 |
+
"step": 35000
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 4.63,
|
| 495 |
+
"learning_rate": 5.8370912391162546e-05,
|
| 496 |
+
"loss": 2.8938,
|
| 497 |
+
"step": 35500
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 4.7,
|
| 501 |
+
"learning_rate": 5.766187356419638e-05,
|
| 502 |
+
"loss": 2.8858,
|
| 503 |
+
"step": 36000
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 4.76,
|
| 507 |
+
"learning_rate": 5.6954252814884144e-05,
|
| 508 |
+
"loss": 2.8869,
|
| 509 |
+
"step": 36500
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 4.83,
|
| 513 |
+
"learning_rate": 5.6245213987917976e-05,
|
| 514 |
+
"loss": 2.883,
|
| 515 |
+
"step": 37000
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 4.89,
|
| 519 |
+
"learning_rate": 5.553617516095182e-05,
|
| 520 |
+
"loss": 2.8871,
|
| 521 |
+
"step": 37500
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 4.96,
|
| 525 |
+
"learning_rate": 5.4827136333985654e-05,
|
| 526 |
+
"loss": 2.8859,
|
| 527 |
+
"step": 38000
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 5.02,
|
| 531 |
+
"learning_rate": 5.411809750701948e-05,
|
| 532 |
+
"loss": 2.8822,
|
| 533 |
+
"step": 38500
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 5.09,
|
| 537 |
+
"learning_rate": 5.341047675770725e-05,
|
| 538 |
+
"loss": 2.8747,
|
| 539 |
+
"step": 39000
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 5.15,
|
| 543 |
+
"learning_rate": 5.270285600839502e-05,
|
| 544 |
+
"loss": 2.8729,
|
| 545 |
+
"step": 39500
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 5.22,
|
| 549 |
+
"learning_rate": 5.1993817181428863e-05,
|
| 550 |
+
"loss": 2.8692,
|
| 551 |
+
"step": 40000
|
| 552 |
+
},
|
| 553 |
+
{
|
| 554 |
+
"epoch": 5.22,
|
| 555 |
+
"eval_loss": 2.8506710529327393,
|
| 556 |
+
"eval_runtime": 8365.1128,
|
| 557 |
+
"eval_samples_per_second": 0.291,
|
| 558 |
+
"eval_steps_per_second": 0.036,
|
| 559 |
+
"eval_wer": 1.3797645737346684,
|
| 560 |
+
"step": 40000
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 5.28,
|
| 564 |
+
"learning_rate": 5.1284778354462696e-05,
|
| 565 |
+
"loss": 2.869,
|
| 566 |
+
"step": 40500
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 5.35,
|
| 570 |
+
"learning_rate": 5.057573952749652e-05,
|
| 571 |
+
"loss": 2.8641,
|
| 572 |
+
"step": 41000
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 5.41,
|
| 576 |
+
"learning_rate": 4.986670070053036e-05,
|
| 577 |
+
"loss": 2.8724,
|
| 578 |
+
"step": 41500
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 5.48,
|
| 582 |
+
"learning_rate": 4.91576618735642e-05,
|
| 583 |
+
"loss": 2.8617,
|
| 584 |
+
"step": 42000
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 5.54,
|
| 588 |
+
"learning_rate": 4.8450041124251965e-05,
|
| 589 |
+
"loss": 2.8586,
|
| 590 |
+
"step": 42500
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 5.61,
|
| 594 |
+
"learning_rate": 4.7741002297285804e-05,
|
| 595 |
+
"loss": 2.8655,
|
| 596 |
+
"step": 43000
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 5.67,
|
| 600 |
+
"learning_rate": 4.703196347031964e-05,
|
| 601 |
+
"loss": 2.866,
|
| 602 |
+
"step": 43500
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 5.74,
|
| 606 |
+
"learning_rate": 4.632292464335347e-05,
|
| 607 |
+
"loss": 2.8702,
|
| 608 |
+
"step": 44000
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 5.81,
|
| 612 |
+
"learning_rate": 4.561388581638731e-05,
|
| 613 |
+
"loss": 2.8603,
|
| 614 |
+
"step": 44500
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 5.87,
|
| 618 |
+
"learning_rate": 4.4906265067075074e-05,
|
| 619 |
+
"loss": 2.8652,
|
| 620 |
+
"step": 45000
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 5.87,
|
| 624 |
+
"eval_loss": 2.8389720916748047,
|
| 625 |
+
"eval_runtime": 8250.6574,
|
| 626 |
+
"eval_samples_per_second": 0.295,
|
| 627 |
+
"eval_steps_per_second": 0.037,
|
| 628 |
+
"eval_wer": 1.4632943395432523,
|
| 629 |
+
"step": 45000
|
| 630 |
+
},
|
| 631 |
+
{
|
| 632 |
+
"epoch": 5.94,
|
| 633 |
+
"learning_rate": 4.419722624010891e-05,
|
| 634 |
+
"loss": 2.8589,
|
| 635 |
+
"step": 45500
|
| 636 |
+
},
|
| 637 |
+
{
|
| 638 |
+
"epoch": 6.0,
|
| 639 |
+
"learning_rate": 4.3488187413142745e-05,
|
| 640 |
+
"loss": 2.8625,
|
| 641 |
+
"step": 46000
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 6.07,
|
| 645 |
+
"learning_rate": 4.2779148586176584e-05,
|
| 646 |
+
"loss": 2.8415,
|
| 647 |
+
"step": 46500
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 6.13,
|
| 651 |
+
"learning_rate": 4.207010975921042e-05,
|
| 652 |
+
"loss": 2.8463,
|
| 653 |
+
"step": 47000
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 6.2,
|
| 657 |
+
"learning_rate": 4.136107093224425e-05,
|
| 658 |
+
"loss": 2.8392,
|
| 659 |
+
"step": 47500
|
| 660 |
+
},
|
| 661 |
+
{
|
| 662 |
+
"epoch": 6.26,
|
| 663 |
+
"learning_rate": 4.065203210527809e-05,
|
| 664 |
+
"loss": 2.8471,
|
| 665 |
+
"step": 48000
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 6.33,
|
| 669 |
+
"learning_rate": 3.994299327831193e-05,
|
| 670 |
+
"loss": 2.8507,
|
| 671 |
+
"step": 48500
|
| 672 |
+
},
|
| 673 |
+
{
|
| 674 |
+
"epoch": 6.39,
|
| 675 |
+
"learning_rate": 3.923395445134575e-05,
|
| 676 |
+
"loss": 2.8451,
|
| 677 |
+
"step": 49000
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"epoch": 6.46,
|
| 681 |
+
"learning_rate": 3.852491562437959e-05,
|
| 682 |
+
"loss": 2.8463,
|
| 683 |
+
"step": 49500
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 6.52,
|
| 687 |
+
"learning_rate": 3.781587679741343e-05,
|
| 688 |
+
"loss": 2.8477,
|
| 689 |
+
"step": 50000
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 6.52,
|
| 693 |
+
"eval_loss": 2.831930160522461,
|
| 694 |
+
"eval_runtime": 8120.4635,
|
| 695 |
+
"eval_samples_per_second": 0.299,
|
| 696 |
+
"eval_steps_per_second": 0.037,
|
| 697 |
+
"eval_wer": 1.360951876558997,
|
| 698 |
+
"step": 50000
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 6.59,
|
| 702 |
+
"learning_rate": 3.71082560481012e-05,
|
| 703 |
+
"loss": 2.8473,
|
| 704 |
+
"step": 50500
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 6.65,
|
| 708 |
+
"learning_rate": 3.639921722113503e-05,
|
| 709 |
+
"loss": 2.8373,
|
| 710 |
+
"step": 51000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 6.72,
|
| 714 |
+
"learning_rate": 3.569017839416887e-05,
|
| 715 |
+
"loss": 2.8431,
|
| 716 |
+
"step": 51500
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 6.78,
|
| 720 |
+
"learning_rate": 3.49811395672027e-05,
|
| 721 |
+
"loss": 2.838,
|
| 722 |
+
"step": 52000
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 6.85,
|
| 726 |
+
"learning_rate": 3.427210074023653e-05,
|
| 727 |
+
"loss": 2.845,
|
| 728 |
+
"step": 52500
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 6.91,
|
| 732 |
+
"learning_rate": 3.356306191327037e-05,
|
| 733 |
+
"loss": 2.8389,
|
| 734 |
+
"step": 53000
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 6.98,
|
| 738 |
+
"learning_rate": 3.285402308630421e-05,
|
| 739 |
+
"loss": 2.8464,
|
| 740 |
+
"step": 53500
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 7.04,
|
| 744 |
+
"learning_rate": 3.214640233699198e-05,
|
| 745 |
+
"loss": 2.8313,
|
| 746 |
+
"step": 54000
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 7.11,
|
| 750 |
+
"learning_rate": 3.143736351002581e-05,
|
| 751 |
+
"loss": 2.8282,
|
| 752 |
+
"step": 54500
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 7.18,
|
| 756 |
+
"learning_rate": 3.072832468305965e-05,
|
| 757 |
+
"loss": 2.8333,
|
| 758 |
+
"step": 55000
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 7.18,
|
| 762 |
+
"eval_loss": 2.822190999984741,
|
| 763 |
+
"eval_runtime": 7683.7145,
|
| 764 |
+
"eval_samples_per_second": 0.317,
|
| 765 |
+
"eval_steps_per_second": 0.04,
|
| 766 |
+
"eval_wer": 1.1936536074515367,
|
| 767 |
+
"step": 55000
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 7.24,
|
| 771 |
+
"learning_rate": 3.001928585609348e-05,
|
| 772 |
+
"loss": 2.8288,
|
| 773 |
+
"step": 55500
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 7.31,
|
| 777 |
+
"learning_rate": 2.931024702912732e-05,
|
| 778 |
+
"loss": 2.8279,
|
| 779 |
+
"step": 56000
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 7.37,
|
| 783 |
+
"learning_rate": 2.860262627981509e-05,
|
| 784 |
+
"loss": 2.8291,
|
| 785 |
+
"step": 56500
|
| 786 |
+
},
|
| 787 |
+
{
|
| 788 |
+
"epoch": 7.44,
|
| 789 |
+
"learning_rate": 2.7893587452848917e-05,
|
| 790 |
+
"loss": 2.8336,
|
| 791 |
+
"step": 57000
|
| 792 |
+
},
|
| 793 |
+
{
|
| 794 |
+
"epoch": 7.5,
|
| 795 |
+
"learning_rate": 2.7184548625882757e-05,
|
| 796 |
+
"loss": 2.8292,
|
| 797 |
+
"step": 57500
|
| 798 |
+
},
|
| 799 |
+
{
|
| 800 |
+
"epoch": 7.57,
|
| 801 |
+
"learning_rate": 2.6475509798916592e-05,
|
| 802 |
+
"loss": 2.831,
|
| 803 |
+
"step": 58000
|
| 804 |
+
},
|
| 805 |
+
{
|
| 806 |
+
"epoch": 7.63,
|
| 807 |
+
"learning_rate": 2.5766470971950425e-05,
|
| 808 |
+
"loss": 2.8265,
|
| 809 |
+
"step": 58500
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 7.7,
|
| 813 |
+
"learning_rate": 2.505743214498426e-05,
|
| 814 |
+
"loss": 2.8229,
|
| 815 |
+
"step": 59000
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 7.76,
|
| 819 |
+
"learning_rate": 2.4348393318018096e-05,
|
| 820 |
+
"loss": 2.8232,
|
| 821 |
+
"step": 59500
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 7.83,
|
| 825 |
+
"learning_rate": 2.364077256870586e-05,
|
| 826 |
+
"loss": 2.8236,
|
| 827 |
+
"step": 60000
|
| 828 |
+
},
|
| 829 |
+
{
|
| 830 |
+
"epoch": 7.83,
|
| 831 |
+
"eval_loss": 2.8168787956237793,
|
| 832 |
+
"eval_runtime": 8146.3496,
|
| 833 |
+
"eval_samples_per_second": 0.299,
|
| 834 |
+
"eval_steps_per_second": 0.037,
|
| 835 |
+
"eval_wer": 1.3519813839639754,
|
| 836 |
+
"step": 60000
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 7.89,
|
| 840 |
+
"learning_rate": 2.29317337417397e-05,
|
| 841 |
+
"loss": 2.8204,
|
| 842 |
+
"step": 60500
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 7.96,
|
| 846 |
+
"learning_rate": 2.2222694914773533e-05,
|
| 847 |
+
"loss": 2.8284,
|
| 848 |
+
"step": 61000
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 8.02,
|
| 852 |
+
"learning_rate": 2.151365608780737e-05,
|
| 853 |
+
"loss": 2.8229,
|
| 854 |
+
"step": 61500
|
| 855 |
+
},
|
| 856 |
+
{
|
| 857 |
+
"epoch": 8.09,
|
| 858 |
+
"learning_rate": 2.0804617260841205e-05,
|
| 859 |
+
"loss": 2.8108,
|
| 860 |
+
"step": 62000
|
| 861 |
+
},
|
| 862 |
+
{
|
| 863 |
+
"epoch": 8.15,
|
| 864 |
+
"learning_rate": 2.009557843387504e-05,
|
| 865 |
+
"loss": 2.8219,
|
| 866 |
+
"step": 62500
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"epoch": 8.22,
|
| 870 |
+
"learning_rate": 1.9386539606908873e-05,
|
| 871 |
+
"loss": 2.816,
|
| 872 |
+
"step": 63000
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 8.28,
|
| 876 |
+
"learning_rate": 1.867891885759664e-05,
|
| 877 |
+
"loss": 2.8139,
|
| 878 |
+
"step": 63500
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 8.35,
|
| 882 |
+
"learning_rate": 1.7969880030630477e-05,
|
| 883 |
+
"loss": 2.819,
|
| 884 |
+
"step": 64000
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 8.41,
|
| 888 |
+
"learning_rate": 1.7260841203664313e-05,
|
| 889 |
+
"loss": 2.8137,
|
| 890 |
+
"step": 64500
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 8.48,
|
| 894 |
+
"learning_rate": 1.655180237669815e-05,
|
| 895 |
+
"loss": 2.8151,
|
| 896 |
+
"step": 65000
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 8.48,
|
| 900 |
+
"eval_loss": 2.811957597732544,
|
| 901 |
+
"eval_runtime": 7992.8193,
|
| 902 |
+
"eval_samples_per_second": 0.304,
|
| 903 |
+
"eval_steps_per_second": 0.038,
|
| 904 |
+
"eval_wer": 1.2878607017052808,
|
| 905 |
+
"step": 65000
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 8.54,
|
| 909 |
+
"learning_rate": 1.5844181627385914e-05,
|
| 910 |
+
"loss": 2.8081,
|
| 911 |
+
"step": 65500
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"epoch": 8.61,
|
| 915 |
+
"learning_rate": 1.5135142800419752e-05,
|
| 916 |
+
"loss": 2.8113,
|
| 917 |
+
"step": 66000
|
| 918 |
+
},
|
| 919 |
+
{
|
| 920 |
+
"epoch": 8.68,
|
| 921 |
+
"learning_rate": 1.4426103973453586e-05,
|
| 922 |
+
"loss": 2.8156,
|
| 923 |
+
"step": 66500
|
| 924 |
+
},
|
| 925 |
+
{
|
| 926 |
+
"epoch": 8.74,
|
| 927 |
+
"learning_rate": 1.3717065146487423e-05,
|
| 928 |
+
"loss": 2.8218,
|
| 929 |
+
"step": 67000
|
| 930 |
+
},
|
| 931 |
+
{
|
| 932 |
+
"epoch": 8.81,
|
| 933 |
+
"learning_rate": 1.3008026319521257e-05,
|
| 934 |
+
"loss": 2.8164,
|
| 935 |
+
"step": 67500
|
| 936 |
+
},
|
| 937 |
+
{
|
| 938 |
+
"epoch": 8.87,
|
| 939 |
+
"learning_rate": 1.2298987492555093e-05,
|
| 940 |
+
"loss": 2.8072,
|
| 941 |
+
"step": 68000
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 8.94,
|
| 945 |
+
"learning_rate": 1.1589948665588929e-05,
|
| 946 |
+
"loss": 2.8064,
|
| 947 |
+
"step": 68500
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 9.0,
|
| 951 |
+
"learning_rate": 1.0880909838622763e-05,
|
| 952 |
+
"loss": 2.8106,
|
| 953 |
+
"step": 69000
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 9.07,
|
| 957 |
+
"learning_rate": 1.0171871011656599e-05,
|
| 958 |
+
"loss": 2.8085,
|
| 959 |
+
"step": 69500
|
| 960 |
+
},
|
| 961 |
+
{
|
| 962 |
+
"epoch": 9.13,
|
| 963 |
+
"learning_rate": 9.462832184690434e-06,
|
| 964 |
+
"loss": 2.8081,
|
| 965 |
+
"step": 70000
|
| 966 |
+
},
|
| 967 |
+
{
|
| 968 |
+
"epoch": 9.13,
|
| 969 |
+
"eval_loss": 2.808976888656616,
|
| 970 |
+
"eval_runtime": 8181.5451,
|
| 971 |
+
"eval_samples_per_second": 0.297,
|
| 972 |
+
"eval_steps_per_second": 0.037,
|
| 973 |
+
"eval_wer": 1.3355642940184824,
|
| 974 |
+
"step": 70000
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 9.2,
|
| 978 |
+
"learning_rate": 8.75379335772427e-06,
|
| 979 |
+
"loss": 2.8072,
|
| 980 |
+
"step": 70500
|
| 981 |
+
},
|
| 982 |
+
{
|
| 983 |
+
"epoch": 9.26,
|
| 984 |
+
"learning_rate": 8.044754530758104e-06,
|
| 985 |
+
"loss": 2.8064,
|
| 986 |
+
"step": 71000
|
| 987 |
+
},
|
| 988 |
+
{
|
| 989 |
+
"epoch": 9.33,
|
| 990 |
+
"learning_rate": 7.33571570379194e-06,
|
| 991 |
+
"loss": 2.8067,
|
| 992 |
+
"step": 71500
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 9.39,
|
| 996 |
+
"learning_rate": 6.628094954479708e-06,
|
| 997 |
+
"loss": 2.8033,
|
| 998 |
+
"step": 72000
|
| 999 |
+
},
|
| 1000 |
+
{
|
| 1001 |
+
"epoch": 9.46,
|
| 1002 |
+
"learning_rate": 5.919056127513543e-06,
|
| 1003 |
+
"loss": 2.8104,
|
| 1004 |
+
"step": 72500
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 9.52,
|
| 1008 |
+
"learning_rate": 5.210017300547379e-06,
|
| 1009 |
+
"loss": 2.8062,
|
| 1010 |
+
"step": 73000
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 9.59,
|
| 1014 |
+
"learning_rate": 4.502396551235146e-06,
|
| 1015 |
+
"loss": 2.7983,
|
| 1016 |
+
"step": 73500
|
| 1017 |
+
},
|
| 1018 |
+
{
|
| 1019 |
+
"epoch": 9.65,
|
| 1020 |
+
"learning_rate": 3.793357724268981e-06,
|
| 1021 |
+
"loss": 2.7966,
|
| 1022 |
+
"step": 74000
|
| 1023 |
+
},
|
| 1024 |
+
{
|
| 1025 |
+
"epoch": 9.72,
|
| 1026 |
+
"learning_rate": 3.0843188973028165e-06,
|
| 1027 |
+
"loss": 2.8003,
|
| 1028 |
+
"step": 74500
|
| 1029 |
+
},
|
| 1030 |
+
{
|
| 1031 |
+
"epoch": 9.78,
|
| 1032 |
+
"learning_rate": 2.375280070336652e-06,
|
| 1033 |
+
"loss": 2.8038,
|
| 1034 |
+
"step": 75000
|
| 1035 |
+
},
|
| 1036 |
+
{
|
| 1037 |
+
"epoch": 9.78,
|
| 1038 |
+
"eval_loss": 2.805771589279175,
|
| 1039 |
+
"eval_runtime": 8283.5766,
|
| 1040 |
+
"eval_samples_per_second": 0.294,
|
| 1041 |
+
"eval_steps_per_second": 0.037,
|
| 1042 |
+
"eval_wer": 1.3321451341000374,
|
| 1043 |
+
"step": 75000
|
| 1044 |
+
}
|
| 1045 |
+
],
|
| 1046 |
+
"max_steps": 76650,
|
| 1047 |
+
"num_train_epochs": 10,
|
| 1048 |
+
"total_flos": 7.309582504377754e+17,
|
| 1049 |
+
"trial_name": null,
|
| 1050 |
+
"trial_params": null
|
| 1051 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c58d3582c52732ca3671408adb737f04572c28a3b0a3e3f288f1a61a5de97c27
|
| 3 |
+
size 3643
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|