Upload 15 files
Browse files- config.json +116 -0
- optimizer.pt +3 -0
- preprocessor_config.json +9 -0
- pytorch_model.bin +3 -0
- rng_state_0.pth +3 -0
- rng_state_1.pth +3 -0
- rng_state_2.pth +3 -0
- rng_state_3.pth +3 -0
- scaler.pt +3 -0
- scheduler.pt +3 -0
- special_tokens_map.json +1 -0
- tokenizer_config.json +1 -0
- trainer_state.json +2161 -0
- training_args.bin +3 -0
- vocab.json +1 -0
config.json
ADDED
|
@@ -0,0 +1,116 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "facebook/wav2vec2-large-xlsr-53",
|
| 3 |
+
"activation_dropout": 0.055,
|
| 4 |
+
"adapter_kernel_size": 3,
|
| 5 |
+
"adapter_stride": 2,
|
| 6 |
+
"add_adapter": false,
|
| 7 |
+
"apply_spec_augment": true,
|
| 8 |
+
"architectures": [
|
| 9 |
+
"Wav2Vec2ForCTC"
|
| 10 |
+
],
|
| 11 |
+
"attention_dropout": 0.094,
|
| 12 |
+
"bos_token_id": 1,
|
| 13 |
+
"classifier_proj_size": 256,
|
| 14 |
+
"codevector_dim": 768,
|
| 15 |
+
"contrastive_logits_temperature": 0.1,
|
| 16 |
+
"conv_bias": true,
|
| 17 |
+
"conv_dim": [
|
| 18 |
+
512,
|
| 19 |
+
512,
|
| 20 |
+
512,
|
| 21 |
+
512,
|
| 22 |
+
512,
|
| 23 |
+
512,
|
| 24 |
+
512
|
| 25 |
+
],
|
| 26 |
+
"conv_kernel": [
|
| 27 |
+
10,
|
| 28 |
+
3,
|
| 29 |
+
3,
|
| 30 |
+
3,
|
| 31 |
+
3,
|
| 32 |
+
2,
|
| 33 |
+
2
|
| 34 |
+
],
|
| 35 |
+
"conv_stride": [
|
| 36 |
+
5,
|
| 37 |
+
2,
|
| 38 |
+
2,
|
| 39 |
+
2,
|
| 40 |
+
2,
|
| 41 |
+
2,
|
| 42 |
+
2
|
| 43 |
+
],
|
| 44 |
+
"ctc_loss_reduction": "mean",
|
| 45 |
+
"ctc_zero_infinity": true,
|
| 46 |
+
"diversity_loss_weight": 0.1,
|
| 47 |
+
"do_stable_layer_norm": true,
|
| 48 |
+
"eos_token_id": 2,
|
| 49 |
+
"feat_extract_activation": "gelu",
|
| 50 |
+
"feat_extract_dropout": 0.0,
|
| 51 |
+
"feat_extract_norm": "layer",
|
| 52 |
+
"feat_proj_dropout": 0.04,
|
| 53 |
+
"feat_quantizer_dropout": 0.0,
|
| 54 |
+
"final_dropout": 0.0,
|
| 55 |
+
"gradient_checkpointing": false,
|
| 56 |
+
"hidden_act": "gelu",
|
| 57 |
+
"hidden_dropout": 0.047,
|
| 58 |
+
"hidden_size": 1024,
|
| 59 |
+
"initializer_range": 0.02,
|
| 60 |
+
"intermediate_size": 4096,
|
| 61 |
+
"layer_norm_eps": 1e-05,
|
| 62 |
+
"layerdrop": 0.041,
|
| 63 |
+
"mask_channel_length": 10,
|
| 64 |
+
"mask_channel_min_space": 1,
|
| 65 |
+
"mask_channel_other": 0.0,
|
| 66 |
+
"mask_channel_prob": 0.0,
|
| 67 |
+
"mask_channel_selection": "static",
|
| 68 |
+
"mask_feature_length": 10,
|
| 69 |
+
"mask_feature_min_masks": 0,
|
| 70 |
+
"mask_feature_prob": 0.0,
|
| 71 |
+
"mask_time_length": 10,
|
| 72 |
+
"mask_time_min_masks": 2,
|
| 73 |
+
"mask_time_min_space": 1,
|
| 74 |
+
"mask_time_other": 0.0,
|
| 75 |
+
"mask_time_prob": 0.4,
|
| 76 |
+
"mask_time_selection": "static",
|
| 77 |
+
"model_type": "wav2vec2",
|
| 78 |
+
"num_adapter_layers": 3,
|
| 79 |
+
"num_attention_heads": 16,
|
| 80 |
+
"num_codevector_groups": 2,
|
| 81 |
+
"num_codevectors_per_group": 320,
|
| 82 |
+
"num_conv_pos_embedding_groups": 16,
|
| 83 |
+
"num_conv_pos_embeddings": 128,
|
| 84 |
+
"num_feat_extract_layers": 7,
|
| 85 |
+
"num_hidden_layers": 24,
|
| 86 |
+
"num_negatives": 100,
|
| 87 |
+
"output_hidden_size": 1024,
|
| 88 |
+
"pad_token_id": 49,
|
| 89 |
+
"proj_codevector_dim": 768,
|
| 90 |
+
"tdnn_dilation": [
|
| 91 |
+
1,
|
| 92 |
+
2,
|
| 93 |
+
3,
|
| 94 |
+
1,
|
| 95 |
+
1
|
| 96 |
+
],
|
| 97 |
+
"tdnn_dim": [
|
| 98 |
+
512,
|
| 99 |
+
512,
|
| 100 |
+
512,
|
| 101 |
+
512,
|
| 102 |
+
1500
|
| 103 |
+
],
|
| 104 |
+
"tdnn_kernel": [
|
| 105 |
+
5,
|
| 106 |
+
3,
|
| 107 |
+
3,
|
| 108 |
+
1,
|
| 109 |
+
1
|
| 110 |
+
],
|
| 111 |
+
"torch_dtype": "float32",
|
| 112 |
+
"transformers_version": "4.15.0",
|
| 113 |
+
"use_weighted_layer_sum": false,
|
| 114 |
+
"vocab_size": 50,
|
| 115 |
+
"xvector_output_dim": 512
|
| 116 |
+
}
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:66b7f01ebdd323529c08470f23ea0af183085cdb7f3d07181b3ebf1f815d0fcc
|
| 3 |
+
size 2490469777
|
preprocessor_config.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_normalize": true,
|
| 3 |
+
"feature_extractor_type": "Wav2Vec2FeatureExtractor",
|
| 4 |
+
"feature_size": 1,
|
| 5 |
+
"padding_side": "right",
|
| 6 |
+
"padding_value": 0.0,
|
| 7 |
+
"return_attention_mask": true,
|
| 8 |
+
"sampling_rate": 16000
|
| 9 |
+
}
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2d858aaf611c7b4fec6f144ed02c674bbb7960ed4d4ebab87b5ea1400b76b8bc
|
| 3 |
+
size 1262128689
|
rng_state_0.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8879005e3d90865108046d941ebaf63394a746cb2f850836231f5f7049a00d0b
|
| 3 |
+
size 14503
|
rng_state_1.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:987ab550feb9defbeffe52e7797e6102e45aa5a9524cb83ca4b15f626bc04c6c
|
| 3 |
+
size 14567
|
rng_state_2.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7bc3fd483601142a3d3c16902c846c3d50f2476c1566b5f3d755754eb8f98838
|
| 3 |
+
size 14567
|
rng_state_3.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:da501940cff99c1f0be42aa7e90265e4bf4be175005e8f34bf2ada34f72ec05e
|
| 3 |
+
size 14567
|
scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:95f5857fb7f607fff30e1fd83c07b0e57e29b5e1ce8580a4d951663d32b18b95
|
| 3 |
+
size 559
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9adc9e415274c6602898879d13ca0149e9c40766734c01f84b8f703f0e88f725
|
| 3 |
+
size 623
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "[UNK]", "pad_token": "[PAD]"}
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"unk_token": "[UNK]", "bos_token": "<s>", "eos_token": "</s>", "pad_token": "[PAD]", "do_lower_case": false, "word_delimiter_token": "|", "tokenizer_class": "Wav2Vec2CTCTokenizer"}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2161 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 158.80044407438245,
|
| 5 |
+
"global_step": 286000,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 1.11,
|
| 12 |
+
"learning_rate": 9.968592592592593e-05,
|
| 13 |
+
"loss": 1.9211,
|
| 14 |
+
"step": 2000
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 1.11,
|
| 18 |
+
"eval_loss": 0.3011658489704132,
|
| 19 |
+
"eval_runtime": 109.7461,
|
| 20 |
+
"eval_samples_per_second": 116.214,
|
| 21 |
+
"eval_steps_per_second": 3.636,
|
| 22 |
+
"eval_wer": 0.4120684023643701,
|
| 23 |
+
"step": 2000
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 2.22,
|
| 27 |
+
"learning_rate": 9.931555555555556e-05,
|
| 28 |
+
"loss": 0.9582,
|
| 29 |
+
"step": 4000
|
| 30 |
+
},
|
| 31 |
+
{
|
| 32 |
+
"epoch": 2.22,
|
| 33 |
+
"eval_loss": 0.24203644692897797,
|
| 34 |
+
"eval_runtime": 109.3338,
|
| 35 |
+
"eval_samples_per_second": 116.652,
|
| 36 |
+
"eval_steps_per_second": 3.649,
|
| 37 |
+
"eval_wer": 0.3336880590881792,
|
| 38 |
+
"step": 4000
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 3.33,
|
| 42 |
+
"learning_rate": 9.894518518518519e-05,
|
| 43 |
+
"loss": 0.8899,
|
| 44 |
+
"step": 6000
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 3.33,
|
| 48 |
+
"eval_loss": 0.21842752397060394,
|
| 49 |
+
"eval_runtime": 109.0188,
|
| 50 |
+
"eval_samples_per_second": 116.989,
|
| 51 |
+
"eval_steps_per_second": 3.66,
|
| 52 |
+
"eval_wer": 0.3000136973311277,
|
| 53 |
+
"step": 6000
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 4.44,
|
| 57 |
+
"learning_rate": 9.857500000000001e-05,
|
| 58 |
+
"loss": 0.8488,
|
| 59 |
+
"step": 8000
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 4.44,
|
| 63 |
+
"eval_loss": 0.20210157334804535,
|
| 64 |
+
"eval_runtime": 108.438,
|
| 65 |
+
"eval_samples_per_second": 117.616,
|
| 66 |
+
"eval_steps_per_second": 3.68,
|
| 67 |
+
"eval_wer": 0.28011042156170646,
|
| 68 |
+
"step": 8000
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 5.55,
|
| 72 |
+
"learning_rate": 9.820481481481482e-05,
|
| 73 |
+
"loss": 0.8243,
|
| 74 |
+
"step": 10000
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 5.55,
|
| 78 |
+
"eval_loss": 0.19518321752548218,
|
| 79 |
+
"eval_runtime": 109.2218,
|
| 80 |
+
"eval_samples_per_second": 116.772,
|
| 81 |
+
"eval_steps_per_second": 3.653,
|
| 82 |
+
"eval_wer": 0.2646956558387508,
|
| 83 |
+
"step": 10000
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 6.66,
|
| 87 |
+
"learning_rate": 9.783462962962964e-05,
|
| 88 |
+
"loss": 0.8051,
|
| 89 |
+
"step": 12000
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 6.66,
|
| 93 |
+
"eval_loss": 0.18829651176929474,
|
| 94 |
+
"eval_runtime": 110.5038,
|
| 95 |
+
"eval_samples_per_second": 115.417,
|
| 96 |
+
"eval_steps_per_second": 3.611,
|
| 97 |
+
"eval_wer": 0.25592936391701526,
|
| 98 |
+
"step": 12000
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 7.77,
|
| 102 |
+
"learning_rate": 9.746444444444445e-05,
|
| 103 |
+
"loss": 0.7875,
|
| 104 |
+
"step": 14000
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 7.77,
|
| 108 |
+
"eval_loss": 0.18309463560581207,
|
| 109 |
+
"eval_runtime": 109.5923,
|
| 110 |
+
"eval_samples_per_second": 116.377,
|
| 111 |
+
"eval_steps_per_second": 3.641,
|
| 112 |
+
"eval_wer": 0.24850119588237154,
|
| 113 |
+
"step": 14000
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 8.88,
|
| 117 |
+
"learning_rate": 9.709407407407408e-05,
|
| 118 |
+
"loss": 0.772,
|
| 119 |
+
"step": 16000
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 8.88,
|
| 123 |
+
"eval_loss": 0.17918168008327484,
|
| 124 |
+
"eval_runtime": 109.8993,
|
| 125 |
+
"eval_samples_per_second": 116.052,
|
| 126 |
+
"eval_steps_per_second": 3.631,
|
| 127 |
+
"eval_wer": 0.24318030955968348,
|
| 128 |
+
"step": 16000
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 9.99,
|
| 132 |
+
"learning_rate": 9.672388888888889e-05,
|
| 133 |
+
"loss": 0.7597,
|
| 134 |
+
"step": 18000
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 9.99,
|
| 138 |
+
"eval_loss": 0.17556481063365936,
|
| 139 |
+
"eval_runtime": 109.4542,
|
| 140 |
+
"eval_samples_per_second": 116.524,
|
| 141 |
+
"eval_steps_per_second": 3.645,
|
| 142 |
+
"eval_wer": 0.23213815338903582,
|
| 143 |
+
"step": 18000
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 11.1,
|
| 147 |
+
"learning_rate": 9.635351851851853e-05,
|
| 148 |
+
"loss": 0.7462,
|
| 149 |
+
"step": 20000
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 11.1,
|
| 153 |
+
"eval_loss": 0.17337526381015778,
|
| 154 |
+
"eval_runtime": 107.9125,
|
| 155 |
+
"eval_samples_per_second": 118.188,
|
| 156 |
+
"eval_steps_per_second": 3.697,
|
| 157 |
+
"eval_wer": 0.22965156096892814,
|
| 158 |
+
"step": 20000
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 12.22,
|
| 162 |
+
"learning_rate": 9.598333333333334e-05,
|
| 163 |
+
"loss": 0.7375,
|
| 164 |
+
"step": 22000
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 12.22,
|
| 168 |
+
"eval_loss": 0.16925489902496338,
|
| 169 |
+
"eval_runtime": 109.8662,
|
| 170 |
+
"eval_samples_per_second": 116.087,
|
| 171 |
+
"eval_steps_per_second": 3.632,
|
| 172 |
+
"eval_wer": 0.22406726443224562,
|
| 173 |
+
"step": 22000
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 13.33,
|
| 177 |
+
"learning_rate": 9.561314814814816e-05,
|
| 178 |
+
"loss": 0.7266,
|
| 179 |
+
"step": 24000
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 13.33,
|
| 183 |
+
"eval_loss": 0.16615571081638336,
|
| 184 |
+
"eval_runtime": 108.8823,
|
| 185 |
+
"eval_samples_per_second": 117.136,
|
| 186 |
+
"eval_steps_per_second": 3.665,
|
| 187 |
+
"eval_wer": 0.22065346805887745,
|
| 188 |
+
"step": 24000
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 14.44,
|
| 192 |
+
"learning_rate": 9.524277777777778e-05,
|
| 193 |
+
"loss": 0.7181,
|
| 194 |
+
"step": 26000
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 14.44,
|
| 198 |
+
"eval_loss": 0.16478531062602997,
|
| 199 |
+
"eval_runtime": 108.8044,
|
| 200 |
+
"eval_samples_per_second": 117.22,
|
| 201 |
+
"eval_steps_per_second": 3.667,
|
| 202 |
+
"eval_wer": 0.21787185619909596,
|
| 203 |
+
"step": 26000
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 15.55,
|
| 207 |
+
"learning_rate": 9.487277777777778e-05,
|
| 208 |
+
"loss": 0.7143,
|
| 209 |
+
"step": 28000
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 15.55,
|
| 213 |
+
"eval_loss": 0.16348408162593842,
|
| 214 |
+
"eval_runtime": 109.4918,
|
| 215 |
+
"eval_samples_per_second": 116.484,
|
| 216 |
+
"eval_steps_per_second": 3.644,
|
| 217 |
+
"eval_wer": 0.21597530265833587,
|
| 218 |
+
"step": 28000
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 16.66,
|
| 222 |
+
"learning_rate": 9.45025925925926e-05,
|
| 223 |
+
"loss": 0.7032,
|
| 224 |
+
"step": 30000
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 16.66,
|
| 228 |
+
"eval_loss": 0.16043423116207123,
|
| 229 |
+
"eval_runtime": 109.1929,
|
| 230 |
+
"eval_samples_per_second": 116.802,
|
| 231 |
+
"eval_steps_per_second": 3.654,
|
| 232 |
+
"eval_wer": 0.21385748453782044,
|
| 233 |
+
"step": 30000
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 17.77,
|
| 237 |
+
"learning_rate": 9.413222222222222e-05,
|
| 238 |
+
"loss": 0.6988,
|
| 239 |
+
"step": 32000
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 17.77,
|
| 243 |
+
"eval_loss": 0.16011235117912292,
|
| 244 |
+
"eval_runtime": 109.5116,
|
| 245 |
+
"eval_samples_per_second": 116.463,
|
| 246 |
+
"eval_steps_per_second": 3.643,
|
| 247 |
+
"eval_wer": 0.21078085323836518,
|
| 248 |
+
"step": 32000
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 18.88,
|
| 252 |
+
"learning_rate": 9.376203703703704e-05,
|
| 253 |
+
"loss": 0.6896,
|
| 254 |
+
"step": 34000
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 18.88,
|
| 258 |
+
"eval_loss": 0.15551629662513733,
|
| 259 |
+
"eval_runtime": 109.5231,
|
| 260 |
+
"eval_samples_per_second": 116.45,
|
| 261 |
+
"eval_steps_per_second": 3.643,
|
| 262 |
+
"eval_wer": 0.20503851057328598,
|
| 263 |
+
"step": 34000
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 19.99,
|
| 267 |
+
"learning_rate": 9.339166666666667e-05,
|
| 268 |
+
"loss": 0.6837,
|
| 269 |
+
"step": 36000
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 19.99,
|
| 273 |
+
"eval_loss": 0.15630744397640228,
|
| 274 |
+
"eval_runtime": 111.1508,
|
| 275 |
+
"eval_samples_per_second": 114.745,
|
| 276 |
+
"eval_steps_per_second": 3.59,
|
| 277 |
+
"eval_wer": 0.20782012243306747,
|
| 278 |
+
"step": 36000
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 21.1,
|
| 282 |
+
"learning_rate": 9.302148148148149e-05,
|
| 283 |
+
"loss": 0.6779,
|
| 284 |
+
"step": 38000
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 21.1,
|
| 288 |
+
"eval_loss": 0.15380945801734924,
|
| 289 |
+
"eval_runtime": 108.7667,
|
| 290 |
+
"eval_samples_per_second": 117.26,
|
| 291 |
+
"eval_steps_per_second": 3.668,
|
| 292 |
+
"eval_wer": 0.2020250977251894,
|
| 293 |
+
"step": 38000
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 22.21,
|
| 297 |
+
"learning_rate": 9.26512962962963e-05,
|
| 298 |
+
"loss": 0.6749,
|
| 299 |
+
"step": 40000
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 22.21,
|
| 303 |
+
"eval_loss": 0.15313279628753662,
|
| 304 |
+
"eval_runtime": 109.8109,
|
| 305 |
+
"eval_samples_per_second": 116.145,
|
| 306 |
+
"eval_steps_per_second": 3.634,
|
| 307 |
+
"eval_wer": 0.20110843018048868,
|
| 308 |
+
"step": 40000
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 23.32,
|
| 312 |
+
"learning_rate": 9.228111111111112e-05,
|
| 313 |
+
"loss": 0.6704,
|
| 314 |
+
"step": 42000
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 23.32,
|
| 318 |
+
"eval_loss": 0.15281446278095245,
|
| 319 |
+
"eval_runtime": 109.3199,
|
| 320 |
+
"eval_samples_per_second": 116.667,
|
| 321 |
+
"eval_steps_per_second": 3.65,
|
| 322 |
+
"eval_wer": 0.2010978937719289,
|
| 323 |
+
"step": 42000
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 24.43,
|
| 327 |
+
"learning_rate": 9.191074074074074e-05,
|
| 328 |
+
"loss": 0.6669,
|
| 329 |
+
"step": 44000
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 24.43,
|
| 333 |
+
"eval_loss": 0.15204770863056183,
|
| 334 |
+
"eval_runtime": 109.4873,
|
| 335 |
+
"eval_samples_per_second": 116.488,
|
| 336 |
+
"eval_steps_per_second": 3.644,
|
| 337 |
+
"eval_wer": 0.19591398076051797,
|
| 338 |
+
"step": 44000
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 25.54,
|
| 342 |
+
"learning_rate": 9.154037037037038e-05,
|
| 343 |
+
"loss": 0.6594,
|
| 344 |
+
"step": 46000
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 25.54,
|
| 348 |
+
"eval_loss": 0.14905157685279846,
|
| 349 |
+
"eval_runtime": 108.5995,
|
| 350 |
+
"eval_samples_per_second": 117.441,
|
| 351 |
+
"eval_steps_per_second": 3.674,
|
| 352 |
+
"eval_wer": 0.19542930596676816,
|
| 353 |
+
"step": 46000
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 26.65,
|
| 357 |
+
"learning_rate": 9.117018518518519e-05,
|
| 358 |
+
"loss": 0.6542,
|
| 359 |
+
"step": 48000
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 26.65,
|
| 363 |
+
"eval_loss": 0.14924176037311554,
|
| 364 |
+
"eval_runtime": 110.1651,
|
| 365 |
+
"eval_samples_per_second": 115.772,
|
| 366 |
+
"eval_steps_per_second": 3.622,
|
| 367 |
+
"eval_wer": 0.19538716033252906,
|
| 368 |
+
"step": 48000
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 27.76,
|
| 372 |
+
"learning_rate": 9.079981481481482e-05,
|
| 373 |
+
"loss": 0.652,
|
| 374 |
+
"step": 50000
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 27.76,
|
| 378 |
+
"eval_loss": 0.14801371097564697,
|
| 379 |
+
"eval_runtime": 109.7651,
|
| 380 |
+
"eval_samples_per_second": 116.194,
|
| 381 |
+
"eval_steps_per_second": 3.635,
|
| 382 |
+
"eval_wer": 0.19368026214584497,
|
| 383 |
+
"step": 50000
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 28.87,
|
| 387 |
+
"learning_rate": 9.042962962962963e-05,
|
| 388 |
+
"loss": 0.6474,
|
| 389 |
+
"step": 52000
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 28.87,
|
| 393 |
+
"eval_loss": 0.14799948036670685,
|
| 394 |
+
"eval_runtime": 110.6825,
|
| 395 |
+
"eval_samples_per_second": 115.231,
|
| 396 |
+
"eval_steps_per_second": 3.605,
|
| 397 |
+
"eval_wer": 0.192394820301552,
|
| 398 |
+
"step": 52000
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 29.98,
|
| 402 |
+
"learning_rate": 9.005944444444445e-05,
|
| 403 |
+
"loss": 0.6443,
|
| 404 |
+
"step": 54000
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 29.98,
|
| 408 |
+
"eval_loss": 0.14905065298080444,
|
| 409 |
+
"eval_runtime": 110.5602,
|
| 410 |
+
"eval_samples_per_second": 115.358,
|
| 411 |
+
"eval_steps_per_second": 3.609,
|
| 412 |
+
"eval_wer": 0.19057202162071035,
|
| 413 |
+
"step": 54000
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 31.09,
|
| 417 |
+
"learning_rate": 8.968907407407408e-05,
|
| 418 |
+
"loss": 0.6407,
|
| 419 |
+
"step": 56000
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 31.09,
|
| 423 |
+
"eval_loss": 0.14624294638633728,
|
| 424 |
+
"eval_runtime": 108.1653,
|
| 425 |
+
"eval_samples_per_second": 117.912,
|
| 426 |
+
"eval_steps_per_second": 3.689,
|
| 427 |
+
"eval_wer": 0.19001359196704212,
|
| 428 |
+
"step": 56000
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 32.2,
|
| 432 |
+
"learning_rate": 8.931907407407408e-05,
|
| 433 |
+
"loss": 0.6349,
|
| 434 |
+
"step": 58000
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 32.2,
|
| 438 |
+
"eval_loss": 0.14455664157867432,
|
| 439 |
+
"eval_runtime": 109.1496,
|
| 440 |
+
"eval_samples_per_second": 116.849,
|
| 441 |
+
"eval_steps_per_second": 3.656,
|
| 442 |
+
"eval_wer": 0.18883351420834693,
|
| 443 |
+
"step": 58000
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 33.31,
|
| 447 |
+
"learning_rate": 8.894870370370371e-05,
|
| 448 |
+
"loss": 0.6337,
|
| 449 |
+
"step": 60000
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 33.31,
|
| 453 |
+
"eval_loss": 0.1445704698562622,
|
| 454 |
+
"eval_runtime": 109.4136,
|
| 455 |
+
"eval_samples_per_second": 116.567,
|
| 456 |
+
"eval_steps_per_second": 3.647,
|
| 457 |
+
"eval_wer": 0.18898102392818383,
|
| 458 |
+
"step": 60000
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 34.43,
|
| 462 |
+
"learning_rate": 8.85787037037037e-05,
|
| 463 |
+
"loss": 0.6296,
|
| 464 |
+
"step": 62000
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 34.43,
|
| 468 |
+
"eval_loss": 0.14853893220424652,
|
| 469 |
+
"eval_runtime": 108.9606,
|
| 470 |
+
"eval_samples_per_second": 117.052,
|
| 471 |
+
"eval_steps_per_second": 3.662,
|
| 472 |
+
"eval_wer": 0.18811703842628202,
|
| 473 |
+
"step": 62000
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 35.54,
|
| 477 |
+
"learning_rate": 8.820833333333334e-05,
|
| 478 |
+
"loss": 0.6249,
|
| 479 |
+
"step": 64000
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 35.54,
|
| 483 |
+
"eval_loss": 0.14442408084869385,
|
| 484 |
+
"eval_runtime": 110.1343,
|
| 485 |
+
"eval_samples_per_second": 115.804,
|
| 486 |
+
"eval_steps_per_second": 3.623,
|
| 487 |
+
"eval_wer": 0.18499826149258763,
|
| 488 |
+
"step": 64000
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 36.65,
|
| 492 |
+
"learning_rate": 8.783814814814815e-05,
|
| 493 |
+
"loss": 0.6222,
|
| 494 |
+
"step": 66000
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 36.65,
|
| 498 |
+
"eval_loss": 0.14410334825515747,
|
| 499 |
+
"eval_runtime": 109.3816,
|
| 500 |
+
"eval_samples_per_second": 116.601,
|
| 501 |
+
"eval_steps_per_second": 3.648,
|
| 502 |
+
"eval_wer": 0.18493504304122896,
|
| 503 |
+
"step": 66000
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 37.76,
|
| 507 |
+
"learning_rate": 8.746796296296297e-05,
|
| 508 |
+
"loss": 0.6173,
|
| 509 |
+
"step": 68000
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 37.76,
|
| 513 |
+
"eval_loss": 0.14380425214767456,
|
| 514 |
+
"eval_runtime": 111.8352,
|
| 515 |
+
"eval_samples_per_second": 114.043,
|
| 516 |
+
"eval_steps_per_second": 3.568,
|
| 517 |
+
"eval_wer": 0.18183733892465415,
|
| 518 |
+
"step": 68000
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 38.87,
|
| 522 |
+
"learning_rate": 8.709759259259259e-05,
|
| 523 |
+
"loss": 0.6149,
|
| 524 |
+
"step": 70000
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 38.87,
|
| 528 |
+
"eval_loss": 0.14080771803855896,
|
| 529 |
+
"eval_runtime": 110.5146,
|
| 530 |
+
"eval_samples_per_second": 115.406,
|
| 531 |
+
"eval_steps_per_second": 3.61,
|
| 532 |
+
"eval_wer": 0.18231147730984418,
|
| 533 |
+
"step": 70000
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 39.98,
|
| 537 |
+
"learning_rate": 8.672740740740741e-05,
|
| 538 |
+
"loss": 0.6114,
|
| 539 |
+
"step": 72000
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 39.98,
|
| 543 |
+
"eval_loss": 0.1435374915599823,
|
| 544 |
+
"eval_runtime": 110.4812,
|
| 545 |
+
"eval_samples_per_second": 115.44,
|
| 546 |
+
"eval_steps_per_second": 3.611,
|
| 547 |
+
"eval_wer": 0.18260649674951795,
|
| 548 |
+
"step": 72000
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 41.09,
|
| 552 |
+
"learning_rate": 8.635722222222222e-05,
|
| 553 |
+
"loss": 0.6089,
|
| 554 |
+
"step": 74000
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 41.09,
|
| 558 |
+
"eval_loss": 0.14298424124717712,
|
| 559 |
+
"eval_runtime": 107.7167,
|
| 560 |
+
"eval_samples_per_second": 118.403,
|
| 561 |
+
"eval_steps_per_second": 3.704,
|
| 562 |
+
"eval_wer": 0.18134212772234456,
|
| 563 |
+
"step": 74000
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 42.2,
|
| 567 |
+
"learning_rate": 8.598722222222223e-05,
|
| 568 |
+
"loss": 0.6066,
|
| 569 |
+
"step": 76000
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 42.2,
|
| 573 |
+
"eval_loss": 0.14788854122161865,
|
| 574 |
+
"eval_runtime": 108.9238,
|
| 575 |
+
"eval_samples_per_second": 117.091,
|
| 576 |
+
"eval_steps_per_second": 3.663,
|
| 577 |
+
"eval_wer": 0.18113139955114899,
|
| 578 |
+
"step": 76000
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 43.31,
|
| 582 |
+
"learning_rate": 8.561703703703704e-05,
|
| 583 |
+
"loss": 0.6012,
|
| 584 |
+
"step": 78000
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 43.31,
|
| 588 |
+
"eval_loss": 0.1424262672662735,
|
| 589 |
+
"eval_runtime": 109.6315,
|
| 590 |
+
"eval_samples_per_second": 116.335,
|
| 591 |
+
"eval_steps_per_second": 3.639,
|
| 592 |
+
"eval_wer": 0.18278561569503418,
|
| 593 |
+
"step": 78000
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 44.42,
|
| 597 |
+
"learning_rate": 8.524666666666667e-05,
|
| 598 |
+
"loss": 0.6009,
|
| 599 |
+
"step": 80000
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 44.42,
|
| 603 |
+
"eval_loss": 0.1408558338880539,
|
| 604 |
+
"eval_runtime": 109.6234,
|
| 605 |
+
"eval_samples_per_second": 116.344,
|
| 606 |
+
"eval_steps_per_second": 3.64,
|
| 607 |
+
"eval_wer": 0.17987756693253537,
|
| 608 |
+
"step": 80000
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 45.53,
|
| 612 |
+
"learning_rate": 8.487648148148148e-05,
|
| 613 |
+
"loss": 0.5976,
|
| 614 |
+
"step": 82000
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 45.53,
|
| 618 |
+
"eval_loss": 0.14050035178661346,
|
| 619 |
+
"eval_runtime": 112.9556,
|
| 620 |
+
"eval_samples_per_second": 112.912,
|
| 621 |
+
"eval_steps_per_second": 3.532,
|
| 622 |
+
"eval_wer": 0.1795298654500627,
|
| 623 |
+
"step": 82000
|
| 624 |
+
},
|
| 625 |
+
{
|
| 626 |
+
"epoch": 46.64,
|
| 627 |
+
"learning_rate": 8.450611111111112e-05,
|
| 628 |
+
"loss": 0.5949,
|
| 629 |
+
"step": 84000
|
| 630 |
+
},
|
| 631 |
+
{
|
| 632 |
+
"epoch": 46.64,
|
| 633 |
+
"eval_loss": 0.14021989703178406,
|
| 634 |
+
"eval_runtime": 109.6871,
|
| 635 |
+
"eval_samples_per_second": 116.276,
|
| 636 |
+
"eval_steps_per_second": 3.638,
|
| 637 |
+
"eval_wer": 0.1760317778082163,
|
| 638 |
+
"step": 84000
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 47.75,
|
| 642 |
+
"learning_rate": 8.413592592592593e-05,
|
| 643 |
+
"loss": 0.5928,
|
| 644 |
+
"step": 86000
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 47.75,
|
| 648 |
+
"eval_loss": 0.14247484505176544,
|
| 649 |
+
"eval_runtime": 109.7686,
|
| 650 |
+
"eval_samples_per_second": 116.19,
|
| 651 |
+
"eval_steps_per_second": 3.635,
|
| 652 |
+
"eval_wer": 0.1776859939521015,
|
| 653 |
+
"step": 86000
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 48.86,
|
| 657 |
+
"learning_rate": 8.376574074074075e-05,
|
| 658 |
+
"loss": 0.5886,
|
| 659 |
+
"step": 88000
|
| 660 |
+
},
|
| 661 |
+
{
|
| 662 |
+
"epoch": 48.86,
|
| 663 |
+
"eval_loss": 0.1418529599905014,
|
| 664 |
+
"eval_runtime": 110.4954,
|
| 665 |
+
"eval_samples_per_second": 115.426,
|
| 666 |
+
"eval_steps_per_second": 3.611,
|
| 667 |
+
"eval_wer": 0.17910840910767156,
|
| 668 |
+
"step": 88000
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 49.97,
|
| 672 |
+
"learning_rate": 8.339555555555556e-05,
|
| 673 |
+
"loss": 0.5848,
|
| 674 |
+
"step": 90000
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 49.97,
|
| 678 |
+
"eval_loss": 0.14168688654899597,
|
| 679 |
+
"eval_runtime": 111.002,
|
| 680 |
+
"eval_samples_per_second": 114.899,
|
| 681 |
+
"eval_steps_per_second": 3.595,
|
| 682 |
+
"eval_wer": 0.1777281395863406,
|
| 683 |
+
"step": 90000
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 51.08,
|
| 687 |
+
"learning_rate": 8.302537037037037e-05,
|
| 688 |
+
"loss": 0.5824,
|
| 689 |
+
"step": 92000
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 51.08,
|
| 693 |
+
"eval_loss": 0.1420871764421463,
|
| 694 |
+
"eval_runtime": 113.8377,
|
| 695 |
+
"eval_samples_per_second": 112.037,
|
| 696 |
+
"eval_steps_per_second": 3.505,
|
| 697 |
+
"eval_wer": 0.17755955704938414,
|
| 698 |
+
"step": 92000
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 52.19,
|
| 702 |
+
"learning_rate": 8.2655e-05,
|
| 703 |
+
"loss": 0.5814,
|
| 704 |
+
"step": 94000
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 52.19,
|
| 708 |
+
"eval_loss": 0.14134614169597626,
|
| 709 |
+
"eval_runtime": 108.8305,
|
| 710 |
+
"eval_samples_per_second": 117.191,
|
| 711 |
+
"eval_steps_per_second": 3.666,
|
| 712 |
+
"eval_wer": 0.17652698901052588,
|
| 713 |
+
"step": 94000
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 53.3,
|
| 717 |
+
"learning_rate": 8.228481481481481e-05,
|
| 718 |
+
"loss": 0.5774,
|
| 719 |
+
"step": 96000
|
| 720 |
+
},
|
| 721 |
+
{
|
| 722 |
+
"epoch": 53.3,
|
| 723 |
+
"eval_loss": 0.1393408626317978,
|
| 724 |
+
"eval_runtime": 108.8087,
|
| 725 |
+
"eval_samples_per_second": 117.215,
|
| 726 |
+
"eval_steps_per_second": 3.667,
|
| 727 |
+
"eval_wer": 0.1742721975787333,
|
| 728 |
+
"step": 96000
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 54.41,
|
| 732 |
+
"learning_rate": 8.191462962962963e-05,
|
| 733 |
+
"loss": 0.573,
|
| 734 |
+
"step": 98000
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 54.41,
|
| 738 |
+
"eval_loss": 0.1408652812242508,
|
| 739 |
+
"eval_runtime": 109.0156,
|
| 740 |
+
"eval_samples_per_second": 116.992,
|
| 741 |
+
"eval_steps_per_second": 3.66,
|
| 742 |
+
"eval_wer": 0.1753574476603905,
|
| 743 |
+
"step": 98000
|
| 744 |
+
},
|
| 745 |
+
{
|
| 746 |
+
"epoch": 55.52,
|
| 747 |
+
"learning_rate": 8.154444444444444e-05,
|
| 748 |
+
"loss": 0.573,
|
| 749 |
+
"step": 100000
|
| 750 |
+
},
|
| 751 |
+
{
|
| 752 |
+
"epoch": 55.52,
|
| 753 |
+
"eval_loss": 0.14187079668045044,
|
| 754 |
+
"eval_runtime": 110.6508,
|
| 755 |
+
"eval_samples_per_second": 115.264,
|
| 756 |
+
"eval_steps_per_second": 3.606,
|
| 757 |
+
"eval_wer": 0.17601070499109672,
|
| 758 |
+
"step": 100000
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 56.64,
|
| 762 |
+
"learning_rate": 8.117407407407408e-05,
|
| 763 |
+
"loss": 0.5687,
|
| 764 |
+
"step": 102000
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 56.64,
|
| 768 |
+
"eval_loss": 0.1391436606645584,
|
| 769 |
+
"eval_runtime": 109.8014,
|
| 770 |
+
"eval_samples_per_second": 116.155,
|
| 771 |
+
"eval_steps_per_second": 3.634,
|
| 772 |
+
"eval_wer": 0.1731658746799566,
|
| 773 |
+
"step": 102000
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 57.75,
|
| 777 |
+
"learning_rate": 8.080388888888889e-05,
|
| 778 |
+
"loss": 0.5648,
|
| 779 |
+
"step": 104000
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 57.75,
|
| 783 |
+
"eval_loss": 0.13951122760772705,
|
| 784 |
+
"eval_runtime": 109.5333,
|
| 785 |
+
"eval_samples_per_second": 116.44,
|
| 786 |
+
"eval_steps_per_second": 3.643,
|
| 787 |
+
"eval_wer": 0.1744197072985702,
|
| 788 |
+
"step": 104000
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 58.86,
|
| 792 |
+
"learning_rate": 8.043370370370371e-05,
|
| 793 |
+
"loss": 0.5623,
|
| 794 |
+
"step": 106000
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 58.86,
|
| 798 |
+
"eval_loss": 0.141478031873703,
|
| 799 |
+
"eval_runtime": 126.0697,
|
| 800 |
+
"eval_samples_per_second": 101.166,
|
| 801 |
+
"eval_steps_per_second": 3.165,
|
| 802 |
+
"eval_wer": 0.17249154453213078,
|
| 803 |
+
"step": 106000
|
| 804 |
+
},
|
| 805 |
+
{
|
| 806 |
+
"epoch": 59.97,
|
| 807 |
+
"learning_rate": 8.006333333333333e-05,
|
| 808 |
+
"loss": 0.5613,
|
| 809 |
+
"step": 108000
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 59.97,
|
| 813 |
+
"eval_loss": 0.140051931142807,
|
| 814 |
+
"eval_runtime": 110.0609,
|
| 815 |
+
"eval_samples_per_second": 115.881,
|
| 816 |
+
"eval_steps_per_second": 3.625,
|
| 817 |
+
"eval_wer": 0.17356625820522817,
|
| 818 |
+
"step": 108000
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 61.08,
|
| 822 |
+
"learning_rate": 7.969333333333335e-05,
|
| 823 |
+
"loss": 0.5567,
|
| 824 |
+
"step": 110000
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 61.08,
|
| 828 |
+
"eval_loss": 0.1388338953256607,
|
| 829 |
+
"eval_runtime": 111.9727,
|
| 830 |
+
"eval_samples_per_second": 113.903,
|
| 831 |
+
"eval_steps_per_second": 3.563,
|
| 832 |
+
"eval_wer": 0.17155380417031052,
|
| 833 |
+
"step": 110000
|
| 834 |
+
},
|
| 835 |
+
{
|
| 836 |
+
"epoch": 62.19,
|
| 837 |
+
"learning_rate": 7.932296296296296e-05,
|
| 838 |
+
"loss": 0.5558,
|
| 839 |
+
"step": 112000
|
| 840 |
+
},
|
| 841 |
+
{
|
| 842 |
+
"epoch": 62.19,
|
| 843 |
+
"eval_loss": 0.1424715220928192,
|
| 844 |
+
"eval_runtime": 117.7299,
|
| 845 |
+
"eval_samples_per_second": 108.333,
|
| 846 |
+
"eval_steps_per_second": 3.389,
|
| 847 |
+
"eval_wer": 0.1722386707266961,
|
| 848 |
+
"step": 112000
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 63.3,
|
| 852 |
+
"learning_rate": 7.895277777777778e-05,
|
| 853 |
+
"loss": 0.5515,
|
| 854 |
+
"step": 114000
|
| 855 |
+
},
|
| 856 |
+
{
|
| 857 |
+
"epoch": 63.3,
|
| 858 |
+
"eval_loss": 0.14163948595523834,
|
| 859 |
+
"eval_runtime": 110.1956,
|
| 860 |
+
"eval_samples_per_second": 115.74,
|
| 861 |
+
"eval_steps_per_second": 3.621,
|
| 862 |
+
"eval_wer": 0.1728181731974839,
|
| 863 |
+
"step": 114000
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 64.41,
|
| 867 |
+
"learning_rate": 7.858240740740741e-05,
|
| 868 |
+
"loss": 0.5517,
|
| 869 |
+
"step": 116000
|
| 870 |
+
},
|
| 871 |
+
{
|
| 872 |
+
"epoch": 64.41,
|
| 873 |
+
"eval_loss": 0.1408870965242386,
|
| 874 |
+
"eval_runtime": 110.3371,
|
| 875 |
+
"eval_samples_per_second": 115.591,
|
| 876 |
+
"eval_steps_per_second": 3.616,
|
| 877 |
+
"eval_wer": 0.17021568028321865,
|
| 878 |
+
"step": 116000
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 65.52,
|
| 882 |
+
"learning_rate": 7.821222222222223e-05,
|
| 883 |
+
"loss": 0.5462,
|
| 884 |
+
"step": 118000
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 65.52,
|
| 888 |
+
"eval_loss": 0.14396576583385468,
|
| 889 |
+
"eval_runtime": 110.3476,
|
| 890 |
+
"eval_samples_per_second": 115.58,
|
| 891 |
+
"eval_steps_per_second": 3.616,
|
| 892 |
+
"eval_wer": 0.1711112750107998,
|
| 893 |
+
"step": 118000
|
| 894 |
+
},
|
| 895 |
+
{
|
| 896 |
+
"epoch": 66.63,
|
| 897 |
+
"learning_rate": 7.784203703703704e-05,
|
| 898 |
+
"loss": 0.5468,
|
| 899 |
+
"step": 120000
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 66.63,
|
| 903 |
+
"eval_loss": 0.14202018082141876,
|
| 904 |
+
"eval_runtime": 110.3323,
|
| 905 |
+
"eval_samples_per_second": 115.596,
|
| 906 |
+
"eval_steps_per_second": 3.616,
|
| 907 |
+
"eval_wer": 0.16947813168403417,
|
| 908 |
+
"step": 120000
|
| 909 |
+
},
|
| 910 |
+
{
|
| 911 |
+
"epoch": 67.74,
|
| 912 |
+
"learning_rate": 7.747166666666667e-05,
|
| 913 |
+
"loss": 0.5436,
|
| 914 |
+
"step": 122000
|
| 915 |
+
},
|
| 916 |
+
{
|
| 917 |
+
"epoch": 67.74,
|
| 918 |
+
"eval_loss": 0.1415812373161316,
|
| 919 |
+
"eval_runtime": 111.4691,
|
| 920 |
+
"eval_samples_per_second": 114.417,
|
| 921 |
+
"eval_steps_per_second": 3.579,
|
| 922 |
+
"eval_wer": 0.16994173366066442,
|
| 923 |
+
"step": 122000
|
| 924 |
+
},
|
| 925 |
+
{
|
| 926 |
+
"epoch": 68.85,
|
| 927 |
+
"learning_rate": 7.710166666666667e-05,
|
| 928 |
+
"loss": 0.5391,
|
| 929 |
+
"step": 124000
|
| 930 |
+
},
|
| 931 |
+
{
|
| 932 |
+
"epoch": 68.85,
|
| 933 |
+
"eval_loss": 0.1431398242712021,
|
| 934 |
+
"eval_runtime": 110.7164,
|
| 935 |
+
"eval_samples_per_second": 115.195,
|
| 936 |
+
"eval_steps_per_second": 3.604,
|
| 937 |
+
"eval_wer": 0.1731131926371577,
|
| 938 |
+
"step": 124000
|
| 939 |
+
},
|
| 940 |
+
{
|
| 941 |
+
"epoch": 69.96,
|
| 942 |
+
"learning_rate": 7.673129629629629e-05,
|
| 943 |
+
"loss": 0.5386,
|
| 944 |
+
"step": 126000
|
| 945 |
+
},
|
| 946 |
+
{
|
| 947 |
+
"epoch": 69.96,
|
| 948 |
+
"eval_loss": 0.13997778296470642,
|
| 949 |
+
"eval_runtime": 115.7532,
|
| 950 |
+
"eval_samples_per_second": 110.183,
|
| 951 |
+
"eval_steps_per_second": 3.447,
|
| 952 |
+
"eval_wer": 0.17011031619762088,
|
| 953 |
+
"step": 126000
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 71.07,
|
| 957 |
+
"learning_rate": 7.636111111111111e-05,
|
| 958 |
+
"loss": 0.5355,
|
| 959 |
+
"step": 128000
|
| 960 |
+
},
|
| 961 |
+
{
|
| 962 |
+
"epoch": 71.07,
|
| 963 |
+
"eval_loss": 0.14404296875,
|
| 964 |
+
"eval_runtime": 108.9746,
|
| 965 |
+
"eval_samples_per_second": 117.036,
|
| 966 |
+
"eval_steps_per_second": 3.661,
|
| 967 |
+
"eval_wer": 0.17013138901474043,
|
| 968 |
+
"step": 128000
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 72.18,
|
| 972 |
+
"learning_rate": 7.599074074074074e-05,
|
| 973 |
+
"loss": 0.5326,
|
| 974 |
+
"step": 130000
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 72.18,
|
| 978 |
+
"eval_loss": 0.14536485075950623,
|
| 979 |
+
"eval_runtime": 110.0456,
|
| 980 |
+
"eval_samples_per_second": 115.897,
|
| 981 |
+
"eval_steps_per_second": 3.626,
|
| 982 |
+
"eval_wer": 0.16906721175020284,
|
| 983 |
+
"step": 130000
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 73.29,
|
| 987 |
+
"learning_rate": 7.562074074074074e-05,
|
| 988 |
+
"loss": 0.5312,
|
| 989 |
+
"step": 132000
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 73.29,
|
| 993 |
+
"eval_loss": 0.14173446595668793,
|
| 994 |
+
"eval_runtime": 111.1917,
|
| 995 |
+
"eval_samples_per_second": 114.703,
|
| 996 |
+
"eval_steps_per_second": 3.588,
|
| 997 |
+
"eval_wer": 0.1693306219641973,
|
| 998 |
+
"step": 132000
|
| 999 |
+
},
|
| 1000 |
+
{
|
| 1001 |
+
"epoch": 74.4,
|
| 1002 |
+
"learning_rate": 7.525037037037037e-05,
|
| 1003 |
+
"loss": 0.5274,
|
| 1004 |
+
"step": 134000
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 74.4,
|
| 1008 |
+
"eval_loss": 0.1436368077993393,
|
| 1009 |
+
"eval_runtime": 109.3262,
|
| 1010 |
+
"eval_samples_per_second": 116.66,
|
| 1011 |
+
"eval_steps_per_second": 3.65,
|
| 1012 |
+
"eval_wer": 0.17104805655944114,
|
| 1013 |
+
"step": 134000
|
| 1014 |
+
},
|
| 1015 |
+
{
|
| 1016 |
+
"epoch": 75.51,
|
| 1017 |
+
"learning_rate": 7.488e-05,
|
| 1018 |
+
"loss": 0.5227,
|
| 1019 |
+
"step": 136000
|
| 1020 |
+
},
|
| 1021 |
+
{
|
| 1022 |
+
"epoch": 75.51,
|
| 1023 |
+
"eval_loss": 0.1463368982076645,
|
| 1024 |
+
"eval_runtime": 110.3294,
|
| 1025 |
+
"eval_samples_per_second": 115.599,
|
| 1026 |
+
"eval_steps_per_second": 3.616,
|
| 1027 |
+
"eval_wer": 0.17241778967221233,
|
| 1028 |
+
"step": 136000
|
| 1029 |
+
},
|
| 1030 |
+
{
|
| 1031 |
+
"epoch": 76.62,
|
| 1032 |
+
"learning_rate": 7.450962962962963e-05,
|
| 1033 |
+
"loss": 0.5245,
|
| 1034 |
+
"step": 138000
|
| 1035 |
+
},
|
| 1036 |
+
{
|
| 1037 |
+
"epoch": 76.62,
|
| 1038 |
+
"eval_loss": 0.14312437176704407,
|
| 1039 |
+
"eval_runtime": 109.8942,
|
| 1040 |
+
"eval_samples_per_second": 116.057,
|
| 1041 |
+
"eval_steps_per_second": 3.631,
|
| 1042 |
+
"eval_wer": 0.16633828193322026,
|
| 1043 |
+
"step": 138000
|
| 1044 |
+
},
|
| 1045 |
+
{
|
| 1046 |
+
"epoch": 77.73,
|
| 1047 |
+
"learning_rate": 7.413962962962963e-05,
|
| 1048 |
+
"loss": 0.5216,
|
| 1049 |
+
"step": 140000
|
| 1050 |
+
},
|
| 1051 |
+
{
|
| 1052 |
+
"epoch": 77.73,
|
| 1053 |
+
"eval_loss": 0.14303582906723022,
|
| 1054 |
+
"eval_runtime": 113.419,
|
| 1055 |
+
"eval_samples_per_second": 112.45,
|
| 1056 |
+
"eval_steps_per_second": 3.518,
|
| 1057 |
+
"eval_wer": 0.1681505442055021,
|
| 1058 |
+
"step": 140000
|
| 1059 |
+
},
|
| 1060 |
+
{
|
| 1061 |
+
"epoch": 78.84,
|
| 1062 |
+
"learning_rate": 7.376925925925926e-05,
|
| 1063 |
+
"loss": 0.5163,
|
| 1064 |
+
"step": 142000
|
| 1065 |
+
},
|
| 1066 |
+
{
|
| 1067 |
+
"epoch": 78.84,
|
| 1068 |
+
"eval_loss": 0.14348630607128143,
|
| 1069 |
+
"eval_runtime": 111.4429,
|
| 1070 |
+
"eval_samples_per_second": 114.444,
|
| 1071 |
+
"eval_steps_per_second": 3.58,
|
| 1072 |
+
"eval_wer": 0.1669915392639265,
|
| 1073 |
+
"step": 142000
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 79.96,
|
| 1077 |
+
"learning_rate": 7.339925925925926e-05,
|
| 1078 |
+
"loss": 0.5159,
|
| 1079 |
+
"step": 144000
|
| 1080 |
+
},
|
| 1081 |
+
{
|
| 1082 |
+
"epoch": 79.96,
|
| 1083 |
+
"eval_loss": 0.14502613246440887,
|
| 1084 |
+
"eval_runtime": 110.5689,
|
| 1085 |
+
"eval_samples_per_second": 115.349,
|
| 1086 |
+
"eval_steps_per_second": 3.609,
|
| 1087 |
+
"eval_wer": 0.16607487171922578,
|
| 1088 |
+
"step": 144000
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 81.07,
|
| 1092 |
+
"learning_rate": 7.302888888888889e-05,
|
| 1093 |
+
"loss": 0.5152,
|
| 1094 |
+
"step": 146000
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 81.07,
|
| 1098 |
+
"eval_loss": 0.14235709607601166,
|
| 1099 |
+
"eval_runtime": 107.4013,
|
| 1100 |
+
"eval_samples_per_second": 118.751,
|
| 1101 |
+
"eval_steps_per_second": 3.715,
|
| 1102 |
+
"eval_wer": 0.16677027468417116,
|
| 1103 |
+
"step": 146000
|
| 1104 |
+
},
|
| 1105 |
+
{
|
| 1106 |
+
"epoch": 82.18,
|
| 1107 |
+
"learning_rate": 7.265870370370371e-05,
|
| 1108 |
+
"loss": 0.5114,
|
| 1109 |
+
"step": 148000
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 82.18,
|
| 1113 |
+
"eval_loss": 0.14669308066368103,
|
| 1114 |
+
"eval_runtime": 108.9248,
|
| 1115 |
+
"eval_samples_per_second": 117.09,
|
| 1116 |
+
"eval_steps_per_second": 3.663,
|
| 1117 |
+
"eval_wer": 0.1700049521120231,
|
| 1118 |
+
"step": 148000
|
| 1119 |
+
},
|
| 1120 |
+
{
|
| 1121 |
+
"epoch": 83.29,
|
| 1122 |
+
"learning_rate": 7.228833333333333e-05,
|
| 1123 |
+
"loss": 0.5102,
|
| 1124 |
+
"step": 150000
|
| 1125 |
+
},
|
| 1126 |
+
{
|
| 1127 |
+
"epoch": 83.29,
|
| 1128 |
+
"eval_loss": 0.14800167083740234,
|
| 1129 |
+
"eval_runtime": 110.9546,
|
| 1130 |
+
"eval_samples_per_second": 114.948,
|
| 1131 |
+
"eval_steps_per_second": 3.596,
|
| 1132 |
+
"eval_wer": 0.16751835969191542,
|
| 1133 |
+
"step": 150000
|
| 1134 |
+
},
|
| 1135 |
+
{
|
| 1136 |
+
"epoch": 84.4,
|
| 1137 |
+
"learning_rate": 7.191833333333333e-05,
|
| 1138 |
+
"loss": 0.5074,
|
| 1139 |
+
"step": 152000
|
| 1140 |
+
},
|
| 1141 |
+
{
|
| 1142 |
+
"epoch": 84.4,
|
| 1143 |
+
"eval_loss": 0.14698636531829834,
|
| 1144 |
+
"eval_runtime": 109.1016,
|
| 1145 |
+
"eval_samples_per_second": 116.9,
|
| 1146 |
+
"eval_steps_per_second": 3.657,
|
| 1147 |
+
"eval_wer": 0.16448387402669926,
|
| 1148 |
+
"step": 152000
|
| 1149 |
+
},
|
| 1150 |
+
{
|
| 1151 |
+
"epoch": 85.51,
|
| 1152 |
+
"learning_rate": 7.154796296296296e-05,
|
| 1153 |
+
"loss": 0.5048,
|
| 1154 |
+
"step": 154000
|
| 1155 |
+
},
|
| 1156 |
+
{
|
| 1157 |
+
"epoch": 85.51,
|
| 1158 |
+
"eval_loss": 0.14937056601047516,
|
| 1159 |
+
"eval_runtime": 120.2598,
|
| 1160 |
+
"eval_samples_per_second": 106.054,
|
| 1161 |
+
"eval_steps_per_second": 3.318,
|
| 1162 |
+
"eval_wer": 0.16732870433783942,
|
| 1163 |
+
"step": 154000
|
| 1164 |
+
},
|
| 1165 |
+
{
|
| 1166 |
+
"epoch": 86.62,
|
| 1167 |
+
"learning_rate": 7.117796296296296e-05,
|
| 1168 |
+
"loss": 0.5052,
|
| 1169 |
+
"step": 156000
|
| 1170 |
+
},
|
| 1171 |
+
{
|
| 1172 |
+
"epoch": 86.62,
|
| 1173 |
+
"eval_loss": 0.14464369416236877,
|
| 1174 |
+
"eval_runtime": 109.6362,
|
| 1175 |
+
"eval_samples_per_second": 116.33,
|
| 1176 |
+
"eval_steps_per_second": 3.639,
|
| 1177 |
+
"eval_wer": 0.16599058045074755,
|
| 1178 |
+
"step": 156000
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 87.73,
|
| 1182 |
+
"learning_rate": 7.080759259259259e-05,
|
| 1183 |
+
"loss": 0.5017,
|
| 1184 |
+
"step": 158000
|
| 1185 |
+
},
|
| 1186 |
+
{
|
| 1187 |
+
"epoch": 87.73,
|
| 1188 |
+
"eval_loss": 0.14508092403411865,
|
| 1189 |
+
"eval_runtime": 111.0679,
|
| 1190 |
+
"eval_samples_per_second": 114.831,
|
| 1191 |
+
"eval_steps_per_second": 3.592,
|
| 1192 |
+
"eval_wer": 0.1682559082910999,
|
| 1193 |
+
"step": 158000
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 88.84,
|
| 1197 |
+
"learning_rate": 7.04374074074074e-05,
|
| 1198 |
+
"loss": 0.5009,
|
| 1199 |
+
"step": 160000
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 88.84,
|
| 1203 |
+
"eval_loss": 0.15082955360412598,
|
| 1204 |
+
"eval_runtime": 117.4553,
|
| 1205 |
+
"eval_samples_per_second": 108.586,
|
| 1206 |
+
"eval_steps_per_second": 3.397,
|
| 1207 |
+
"eval_wer": 0.16449441043525903,
|
| 1208 |
+
"step": 160000
|
| 1209 |
+
},
|
| 1210 |
+
{
|
| 1211 |
+
"epoch": 89.95,
|
| 1212 |
+
"learning_rate": 7.006703703703704e-05,
|
| 1213 |
+
"loss": 0.4991,
|
| 1214 |
+
"step": 162000
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 89.95,
|
| 1218 |
+
"eval_loss": 0.14789538085460663,
|
| 1219 |
+
"eval_runtime": 110.7552,
|
| 1220 |
+
"eval_samples_per_second": 115.155,
|
| 1221 |
+
"eval_steps_per_second": 3.603,
|
| 1222 |
+
"eval_wer": 0.16420992740414503,
|
| 1223 |
+
"step": 162000
|
| 1224 |
+
},
|
| 1225 |
+
{
|
| 1226 |
+
"epoch": 91.06,
|
| 1227 |
+
"learning_rate": 6.969666666666666e-05,
|
| 1228 |
+
"loss": 0.4955,
|
| 1229 |
+
"step": 164000
|
| 1230 |
+
},
|
| 1231 |
+
{
|
| 1232 |
+
"epoch": 91.06,
|
| 1233 |
+
"eval_loss": 0.15189045667648315,
|
| 1234 |
+
"eval_runtime": 110.2282,
|
| 1235 |
+
"eval_samples_per_second": 115.705,
|
| 1236 |
+
"eval_steps_per_second": 3.62,
|
| 1237 |
+
"eval_wer": 0.1638832987387919,
|
| 1238 |
+
"step": 164000
|
| 1239 |
+
},
|
| 1240 |
+
{
|
| 1241 |
+
"epoch": 92.17,
|
| 1242 |
+
"learning_rate": 6.932666666666667e-05,
|
| 1243 |
+
"loss": 0.4954,
|
| 1244 |
+
"step": 166000
|
| 1245 |
+
},
|
| 1246 |
+
{
|
| 1247 |
+
"epoch": 92.17,
|
| 1248 |
+
"eval_loss": 0.1472417414188385,
|
| 1249 |
+
"eval_runtime": 112.6524,
|
| 1250 |
+
"eval_samples_per_second": 113.216,
|
| 1251 |
+
"eval_steps_per_second": 3.542,
|
| 1252 |
+
"eval_wer": 0.16437850994110148,
|
| 1253 |
+
"step": 166000
|
| 1254 |
+
},
|
| 1255 |
+
{
|
| 1256 |
+
"epoch": 93.28,
|
| 1257 |
+
"learning_rate": 6.895629629629629e-05,
|
| 1258 |
+
"loss": 0.4937,
|
| 1259 |
+
"step": 168000
|
| 1260 |
+
},
|
| 1261 |
+
{
|
| 1262 |
+
"epoch": 93.28,
|
| 1263 |
+
"eval_loss": 0.14937187731266022,
|
| 1264 |
+
"eval_runtime": 111.7122,
|
| 1265 |
+
"eval_samples_per_second": 114.168,
|
| 1266 |
+
"eval_steps_per_second": 3.572,
|
| 1267 |
+
"eval_wer": 0.16531625030292174,
|
| 1268 |
+
"step": 168000
|
| 1269 |
+
},
|
| 1270 |
+
{
|
| 1271 |
+
"epoch": 94.39,
|
| 1272 |
+
"learning_rate": 6.858611111111111e-05,
|
| 1273 |
+
"loss": 0.4913,
|
| 1274 |
+
"step": 170000
|
| 1275 |
+
},
|
| 1276 |
+
{
|
| 1277 |
+
"epoch": 94.39,
|
| 1278 |
+
"eval_loss": 0.14994442462921143,
|
| 1279 |
+
"eval_runtime": 111.1265,
|
| 1280 |
+
"eval_samples_per_second": 114.77,
|
| 1281 |
+
"eval_steps_per_second": 3.591,
|
| 1282 |
+
"eval_wer": 0.16276643943145538,
|
| 1283 |
+
"step": 170000
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 95.5,
|
| 1287 |
+
"learning_rate": 6.821574074074074e-05,
|
| 1288 |
+
"loss": 0.4877,
|
| 1289 |
+
"step": 172000
|
| 1290 |
+
},
|
| 1291 |
+
{
|
| 1292 |
+
"epoch": 95.5,
|
| 1293 |
+
"eval_loss": 0.15009111166000366,
|
| 1294 |
+
"eval_runtime": 110.8815,
|
| 1295 |
+
"eval_samples_per_second": 115.024,
|
| 1296 |
+
"eval_steps_per_second": 3.598,
|
| 1297 |
+
"eval_wer": 0.16477889346637306,
|
| 1298 |
+
"step": 172000
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 96.61,
|
| 1302 |
+
"learning_rate": 6.784555555555556e-05,
|
| 1303 |
+
"loss": 0.4855,
|
| 1304 |
+
"step": 174000
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 96.61,
|
| 1308 |
+
"eval_loss": 0.15036821365356445,
|
| 1309 |
+
"eval_runtime": 123.2578,
|
| 1310 |
+
"eval_samples_per_second": 103.474,
|
| 1311 |
+
"eval_steps_per_second": 3.237,
|
| 1312 |
+
"eval_wer": 0.16344076957928122,
|
| 1313 |
+
"step": 174000
|
| 1314 |
+
},
|
| 1315 |
+
{
|
| 1316 |
+
"epoch": 97.72,
|
| 1317 |
+
"learning_rate": 6.747537037037037e-05,
|
| 1318 |
+
"loss": 0.4865,
|
| 1319 |
+
"step": 176000
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 97.72,
|
| 1323 |
+
"eval_loss": 0.15151378512382507,
|
| 1324 |
+
"eval_runtime": 110.1192,
|
| 1325 |
+
"eval_samples_per_second": 115.82,
|
| 1326 |
+
"eval_steps_per_second": 3.623,
|
| 1327 |
+
"eval_wer": 0.16542161438851952,
|
| 1328 |
+
"step": 176000
|
| 1329 |
+
},
|
| 1330 |
+
{
|
| 1331 |
+
"epoch": 98.83,
|
| 1332 |
+
"learning_rate": 6.7105e-05,
|
| 1333 |
+
"loss": 0.4804,
|
| 1334 |
+
"step": 178000
|
| 1335 |
+
},
|
| 1336 |
+
{
|
| 1337 |
+
"epoch": 98.83,
|
| 1338 |
+
"eval_loss": 0.1545591503381729,
|
| 1339 |
+
"eval_runtime": 109.0975,
|
| 1340 |
+
"eval_samples_per_second": 116.905,
|
| 1341 |
+
"eval_steps_per_second": 3.657,
|
| 1342 |
+
"eval_wer": 0.16173387139259712,
|
| 1343 |
+
"step": 178000
|
| 1344 |
+
},
|
| 1345 |
+
{
|
| 1346 |
+
"epoch": 99.94,
|
| 1347 |
+
"learning_rate": 6.6735e-05,
|
| 1348 |
+
"loss": 0.4804,
|
| 1349 |
+
"step": 180000
|
| 1350 |
+
},
|
| 1351 |
+
{
|
| 1352 |
+
"epoch": 99.94,
|
| 1353 |
+
"eval_loss": 0.14692111313343048,
|
| 1354 |
+
"eval_runtime": 110.3968,
|
| 1355 |
+
"eval_samples_per_second": 115.529,
|
| 1356 |
+
"eval_steps_per_second": 3.614,
|
| 1357 |
+
"eval_wer": 0.16394651719015058,
|
| 1358 |
+
"step": 180000
|
| 1359 |
+
},
|
| 1360 |
+
{
|
| 1361 |
+
"epoch": 101.05,
|
| 1362 |
+
"learning_rate": 6.636481481481481e-05,
|
| 1363 |
+
"loss": 0.4784,
|
| 1364 |
+
"step": 182000
|
| 1365 |
+
},
|
| 1366 |
+
{
|
| 1367 |
+
"epoch": 101.05,
|
| 1368 |
+
"eval_loss": 0.15785107016563416,
|
| 1369 |
+
"eval_runtime": 119.3969,
|
| 1370 |
+
"eval_samples_per_second": 106.82,
|
| 1371 |
+
"eval_steps_per_second": 3.342,
|
| 1372 |
+
"eval_wer": 0.16587467995659,
|
| 1373 |
+
"step": 182000
|
| 1374 |
+
},
|
| 1375 |
+
{
|
| 1376 |
+
"epoch": 102.17,
|
| 1377 |
+
"learning_rate": 6.599462962962963e-05,
|
| 1378 |
+
"loss": 0.4754,
|
| 1379 |
+
"step": 184000
|
| 1380 |
+
},
|
| 1381 |
+
{
|
| 1382 |
+
"epoch": 102.17,
|
| 1383 |
+
"eval_loss": 0.1541442722082138,
|
| 1384 |
+
"eval_runtime": 110.3894,
|
| 1385 |
+
"eval_samples_per_second": 115.536,
|
| 1386 |
+
"eval_steps_per_second": 3.614,
|
| 1387 |
+
"eval_wer": 0.16502123086324796,
|
| 1388 |
+
"step": 184000
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 103.28,
|
| 1392 |
+
"learning_rate": 6.562425925925926e-05,
|
| 1393 |
+
"loss": 0.4736,
|
| 1394 |
+
"step": 186000
|
| 1395 |
+
},
|
| 1396 |
+
{
|
| 1397 |
+
"epoch": 103.28,
|
| 1398 |
+
"eval_loss": 0.15858229994773865,
|
| 1399 |
+
"eval_runtime": 111.9719,
|
| 1400 |
+
"eval_samples_per_second": 113.904,
|
| 1401 |
+
"eval_steps_per_second": 3.563,
|
| 1402 |
+
"eval_wer": 0.16279804865713474,
|
| 1403 |
+
"step": 186000
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 104.39,
|
| 1407 |
+
"learning_rate": 6.525407407407407e-05,
|
| 1408 |
+
"loss": 0.4733,
|
| 1409 |
+
"step": 188000
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 104.39,
|
| 1413 |
+
"eval_loss": 0.15577121078968048,
|
| 1414 |
+
"eval_runtime": 111.8331,
|
| 1415 |
+
"eval_samples_per_second": 114.045,
|
| 1416 |
+
"eval_steps_per_second": 3.568,
|
| 1417 |
+
"eval_wer": 0.16437850994110148,
|
| 1418 |
+
"step": 188000
|
| 1419 |
+
},
|
| 1420 |
+
{
|
| 1421 |
+
"epoch": 105.5,
|
| 1422 |
+
"learning_rate": 6.48837037037037e-05,
|
| 1423 |
+
"loss": 0.4729,
|
| 1424 |
+
"step": 190000
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 105.5,
|
| 1428 |
+
"eval_loss": 0.15798160433769226,
|
| 1429 |
+
"eval_runtime": 110.7044,
|
| 1430 |
+
"eval_samples_per_second": 115.208,
|
| 1431 |
+
"eval_steps_per_second": 3.604,
|
| 1432 |
+
"eval_wer": 0.16497908522900884,
|
| 1433 |
+
"step": 190000
|
| 1434 |
+
},
|
| 1435 |
+
{
|
| 1436 |
+
"epoch": 106.61,
|
| 1437 |
+
"learning_rate": 6.451351851851852e-05,
|
| 1438 |
+
"loss": 0.4708,
|
| 1439 |
+
"step": 192000
|
| 1440 |
+
},
|
| 1441 |
+
{
|
| 1442 |
+
"epoch": 106.61,
|
| 1443 |
+
"eval_loss": 0.1557486653327942,
|
| 1444 |
+
"eval_runtime": 112.6241,
|
| 1445 |
+
"eval_samples_per_second": 113.244,
|
| 1446 |
+
"eval_steps_per_second": 3.543,
|
| 1447 |
+
"eval_wer": 0.16445226480101993,
|
| 1448 |
+
"step": 192000
|
| 1449 |
+
},
|
| 1450 |
+
{
|
| 1451 |
+
"epoch": 107.72,
|
| 1452 |
+
"learning_rate": 6.414314814814814e-05,
|
| 1453 |
+
"loss": 0.4674,
|
| 1454 |
+
"step": 194000
|
| 1455 |
+
},
|
| 1456 |
+
{
|
| 1457 |
+
"epoch": 107.72,
|
| 1458 |
+
"eval_loss": 0.15497049689292908,
|
| 1459 |
+
"eval_runtime": 111.8369,
|
| 1460 |
+
"eval_samples_per_second": 114.041,
|
| 1461 |
+
"eval_steps_per_second": 3.568,
|
| 1462 |
+
"eval_wer": 0.1644101191667808,
|
| 1463 |
+
"step": 194000
|
| 1464 |
+
},
|
| 1465 |
+
{
|
| 1466 |
+
"epoch": 108.83,
|
| 1467 |
+
"learning_rate": 6.377314814814816e-05,
|
| 1468 |
+
"loss": 0.4645,
|
| 1469 |
+
"step": 196000
|
| 1470 |
+
},
|
| 1471 |
+
{
|
| 1472 |
+
"epoch": 108.83,
|
| 1473 |
+
"eval_loss": 0.15774163603782654,
|
| 1474 |
+
"eval_runtime": 115.5016,
|
| 1475 |
+
"eval_samples_per_second": 110.423,
|
| 1476 |
+
"eval_steps_per_second": 3.454,
|
| 1477 |
+
"eval_wer": 0.16043789313974438,
|
| 1478 |
+
"step": 196000
|
| 1479 |
+
},
|
| 1480 |
+
{
|
| 1481 |
+
"epoch": 109.94,
|
| 1482 |
+
"learning_rate": 6.340277777777777e-05,
|
| 1483 |
+
"loss": 0.4652,
|
| 1484 |
+
"step": 198000
|
| 1485 |
+
},
|
| 1486 |
+
{
|
| 1487 |
+
"epoch": 109.94,
|
| 1488 |
+
"eval_loss": 0.1597561538219452,
|
| 1489 |
+
"eval_runtime": 112.242,
|
| 1490 |
+
"eval_samples_per_second": 113.63,
|
| 1491 |
+
"eval_steps_per_second": 3.555,
|
| 1492 |
+
"eval_wer": 0.16909882097588216,
|
| 1493 |
+
"step": 198000
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 111.05,
|
| 1497 |
+
"learning_rate": 6.303240740740741e-05,
|
| 1498 |
+
"loss": 0.4636,
|
| 1499 |
+
"step": 200000
|
| 1500 |
+
},
|
| 1501 |
+
{
|
| 1502 |
+
"epoch": 111.05,
|
| 1503 |
+
"eval_loss": 0.15712451934814453,
|
| 1504 |
+
"eval_runtime": 116.0757,
|
| 1505 |
+
"eval_samples_per_second": 109.877,
|
| 1506 |
+
"eval_steps_per_second": 3.437,
|
| 1507 |
+
"eval_wer": 0.16316682295672696,
|
| 1508 |
+
"step": 200000
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 112.16,
|
| 1512 |
+
"learning_rate": 6.266222222222222e-05,
|
| 1513 |
+
"loss": 0.4615,
|
| 1514 |
+
"step": 202000
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 112.16,
|
| 1518 |
+
"eval_loss": 0.16025520861148834,
|
| 1519 |
+
"eval_runtime": 109.4143,
|
| 1520 |
+
"eval_samples_per_second": 116.566,
|
| 1521 |
+
"eval_steps_per_second": 3.647,
|
| 1522 |
+
"eval_wer": 0.15968980813200012,
|
| 1523 |
+
"step": 202000
|
| 1524 |
+
},
|
| 1525 |
+
{
|
| 1526 |
+
"epoch": 113.27,
|
| 1527 |
+
"learning_rate": 6.229203703703704e-05,
|
| 1528 |
+
"loss": 0.4586,
|
| 1529 |
+
"step": 204000
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 113.27,
|
| 1533 |
+
"eval_loss": 0.16139160096645355,
|
| 1534 |
+
"eval_runtime": 110.5104,
|
| 1535 |
+
"eval_samples_per_second": 115.41,
|
| 1536 |
+
"eval_steps_per_second": 3.611,
|
| 1537 |
+
"eval_wer": 0.1633986239450421,
|
| 1538 |
+
"step": 204000
|
| 1539 |
+
},
|
| 1540 |
+
{
|
| 1541 |
+
"epoch": 114.38,
|
| 1542 |
+
"learning_rate": 6.192185185185185e-05,
|
| 1543 |
+
"loss": 0.4546,
|
| 1544 |
+
"step": 206000
|
| 1545 |
+
},
|
| 1546 |
+
{
|
| 1547 |
+
"epoch": 114.38,
|
| 1548 |
+
"eval_loss": 0.15872280299663544,
|
| 1549 |
+
"eval_runtime": 111.2156,
|
| 1550 |
+
"eval_samples_per_second": 114.678,
|
| 1551 |
+
"eval_steps_per_second": 3.588,
|
| 1552 |
+
"eval_wer": 0.1650001580461284,
|
| 1553 |
+
"step": 206000
|
| 1554 |
+
},
|
| 1555 |
+
{
|
| 1556 |
+
"epoch": 115.49,
|
| 1557 |
+
"learning_rate": 6.155166666666668e-05,
|
| 1558 |
+
"loss": 0.4554,
|
| 1559 |
+
"step": 208000
|
| 1560 |
+
},
|
| 1561 |
+
{
|
| 1562 |
+
"epoch": 115.49,
|
| 1563 |
+
"eval_loss": 0.1584981381893158,
|
| 1564 |
+
"eval_runtime": 109.0033,
|
| 1565 |
+
"eval_samples_per_second": 117.006,
|
| 1566 |
+
"eval_steps_per_second": 3.66,
|
| 1567 |
+
"eval_wer": 0.16186030829531445,
|
| 1568 |
+
"step": 208000
|
| 1569 |
+
},
|
| 1570 |
+
{
|
| 1571 |
+
"epoch": 116.6,
|
| 1572 |
+
"learning_rate": 6.118148148148148e-05,
|
| 1573 |
+
"loss": 0.4534,
|
| 1574 |
+
"step": 210000
|
| 1575 |
+
},
|
| 1576 |
+
{
|
| 1577 |
+
"epoch": 116.6,
|
| 1578 |
+
"eval_loss": 0.16507840156555176,
|
| 1579 |
+
"eval_runtime": 111.2078,
|
| 1580 |
+
"eval_samples_per_second": 114.686,
|
| 1581 |
+
"eval_steps_per_second": 3.588,
|
| 1582 |
+
"eval_wer": 0.16058540285958128,
|
| 1583 |
+
"step": 210000
|
| 1584 |
+
},
|
| 1585 |
+
{
|
| 1586 |
+
"epoch": 117.71,
|
| 1587 |
+
"learning_rate": 6.081111111111112e-05,
|
| 1588 |
+
"loss": 0.452,
|
| 1589 |
+
"step": 212000
|
| 1590 |
+
},
|
| 1591 |
+
{
|
| 1592 |
+
"epoch": 117.71,
|
| 1593 |
+
"eval_loss": 0.16277721524238586,
|
| 1594 |
+
"eval_runtime": 110.743,
|
| 1595 |
+
"eval_samples_per_second": 115.168,
|
| 1596 |
+
"eval_steps_per_second": 3.603,
|
| 1597 |
+
"eval_wer": 0.16030091982846728,
|
| 1598 |
+
"step": 212000
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 118.82,
|
| 1602 |
+
"learning_rate": 6.044092592592593e-05,
|
| 1603 |
+
"loss": 0.4498,
|
| 1604 |
+
"step": 214000
|
| 1605 |
+
},
|
| 1606 |
+
{
|
| 1607 |
+
"epoch": 118.82,
|
| 1608 |
+
"eval_loss": 0.1640331745147705,
|
| 1609 |
+
"eval_runtime": 111.5522,
|
| 1610 |
+
"eval_samples_per_second": 114.332,
|
| 1611 |
+
"eval_steps_per_second": 3.577,
|
| 1612 |
+
"eval_wer": 0.1609752499762931,
|
| 1613 |
+
"step": 214000
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 119.93,
|
| 1617 |
+
"learning_rate": 6.0070555555555565e-05,
|
| 1618 |
+
"loss": 0.4475,
|
| 1619 |
+
"step": 216000
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 119.93,
|
| 1623 |
+
"eval_loss": 0.16334258019924164,
|
| 1624 |
+
"eval_runtime": 110.4588,
|
| 1625 |
+
"eval_samples_per_second": 115.464,
|
| 1626 |
+
"eval_steps_per_second": 3.612,
|
| 1627 |
+
"eval_wer": 0.16188138111243403,
|
| 1628 |
+
"step": 216000
|
| 1629 |
+
},
|
| 1630 |
+
{
|
| 1631 |
+
"epoch": 121.04,
|
| 1632 |
+
"learning_rate": 5.9700370370370374e-05,
|
| 1633 |
+
"loss": 0.4456,
|
| 1634 |
+
"step": 218000
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 121.04,
|
| 1638 |
+
"eval_loss": 0.1660892814397812,
|
| 1639 |
+
"eval_runtime": 109.1417,
|
| 1640 |
+
"eval_samples_per_second": 116.857,
|
| 1641 |
+
"eval_steps_per_second": 3.656,
|
| 1642 |
+
"eval_wer": 0.15847812114762563,
|
| 1643 |
+
"step": 218000
|
| 1644 |
+
},
|
| 1645 |
+
{
|
| 1646 |
+
"epoch": 122.15,
|
| 1647 |
+
"learning_rate": 5.933018518518518e-05,
|
| 1648 |
+
"loss": 0.4453,
|
| 1649 |
+
"step": 220000
|
| 1650 |
+
},
|
| 1651 |
+
{
|
| 1652 |
+
"epoch": 122.15,
|
| 1653 |
+
"eval_loss": 0.16312651336193085,
|
| 1654 |
+
"eval_runtime": 109.8128,
|
| 1655 |
+
"eval_samples_per_second": 116.143,
|
| 1656 |
+
"eval_steps_per_second": 3.633,
|
| 1657 |
+
"eval_wer": 0.15943693432656544,
|
| 1658 |
+
"step": 220000
|
| 1659 |
+
},
|
| 1660 |
+
{
|
| 1661 |
+
"epoch": 123.26,
|
| 1662 |
+
"learning_rate": 5.8960000000000005e-05,
|
| 1663 |
+
"loss": 0.443,
|
| 1664 |
+
"step": 222000
|
| 1665 |
+
},
|
| 1666 |
+
{
|
| 1667 |
+
"epoch": 123.26,
|
| 1668 |
+
"eval_loss": 0.16088205575942993,
|
| 1669 |
+
"eval_runtime": 110.2399,
|
| 1670 |
+
"eval_samples_per_second": 115.693,
|
| 1671 |
+
"eval_steps_per_second": 3.619,
|
| 1672 |
+
"eval_wer": 0.15721375212045222,
|
| 1673 |
+
"step": 222000
|
| 1674 |
+
},
|
| 1675 |
+
{
|
| 1676 |
+
"epoch": 124.38,
|
| 1677 |
+
"learning_rate": 5.8589814814814813e-05,
|
| 1678 |
+
"loss": 0.4437,
|
| 1679 |
+
"step": 224000
|
| 1680 |
+
},
|
| 1681 |
+
{
|
| 1682 |
+
"epoch": 124.38,
|
| 1683 |
+
"eval_loss": 0.1613088697195053,
|
| 1684 |
+
"eval_runtime": 111.1211,
|
| 1685 |
+
"eval_samples_per_second": 114.776,
|
| 1686 |
+
"eval_steps_per_second": 3.591,
|
| 1687 |
+
"eval_wer": 0.15813041966515293,
|
| 1688 |
+
"step": 224000
|
| 1689 |
+
},
|
| 1690 |
+
{
|
| 1691 |
+
"epoch": 125.49,
|
| 1692 |
+
"learning_rate": 5.821944444444445e-05,
|
| 1693 |
+
"loss": 0.4391,
|
| 1694 |
+
"step": 226000
|
| 1695 |
+
},
|
| 1696 |
+
{
|
| 1697 |
+
"epoch": 125.49,
|
| 1698 |
+
"eval_loss": 0.17190876603126526,
|
| 1699 |
+
"eval_runtime": 125.7786,
|
| 1700 |
+
"eval_samples_per_second": 101.4,
|
| 1701 |
+
"eval_steps_per_second": 3.172,
|
| 1702 |
+
"eval_wer": 0.1587309949530603,
|
| 1703 |
+
"step": 226000
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 126.6,
|
| 1707 |
+
"learning_rate": 5.784925925925926e-05,
|
| 1708 |
+
"loss": 0.4372,
|
| 1709 |
+
"step": 228000
|
| 1710 |
+
},
|
| 1711 |
+
{
|
| 1712 |
+
"epoch": 126.6,
|
| 1713 |
+
"eval_loss": 0.16242261230945587,
|
| 1714 |
+
"eval_runtime": 111.445,
|
| 1715 |
+
"eval_samples_per_second": 114.442,
|
| 1716 |
+
"eval_steps_per_second": 3.58,
|
| 1717 |
+
"eval_wer": 0.15744555310876734,
|
| 1718 |
+
"step": 228000
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 127.71,
|
| 1722 |
+
"learning_rate": 5.747907407407408e-05,
|
| 1723 |
+
"loss": 0.4377,
|
| 1724 |
+
"step": 230000
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 127.71,
|
| 1728 |
+
"eval_loss": 0.16486689448356628,
|
| 1729 |
+
"eval_runtime": 111.8011,
|
| 1730 |
+
"eval_samples_per_second": 114.078,
|
| 1731 |
+
"eval_steps_per_second": 3.569,
|
| 1732 |
+
"eval_wer": 0.15686605063797954,
|
| 1733 |
+
"step": 230000
|
| 1734 |
+
},
|
| 1735 |
+
{
|
| 1736 |
+
"epoch": 128.82,
|
| 1737 |
+
"learning_rate": 5.71087037037037e-05,
|
| 1738 |
+
"loss": 0.4349,
|
| 1739 |
+
"step": 232000
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 128.82,
|
| 1743 |
+
"eval_loss": 0.1673368662595749,
|
| 1744 |
+
"eval_runtime": 112.5569,
|
| 1745 |
+
"eval_samples_per_second": 113.312,
|
| 1746 |
+
"eval_steps_per_second": 3.545,
|
| 1747 |
+
"eval_wer": 0.15947907996080457,
|
| 1748 |
+
"step": 232000
|
| 1749 |
+
},
|
| 1750 |
+
{
|
| 1751 |
+
"epoch": 129.93,
|
| 1752 |
+
"learning_rate": 5.6738518518518525e-05,
|
| 1753 |
+
"loss": 0.4342,
|
| 1754 |
+
"step": 234000
|
| 1755 |
+
},
|
| 1756 |
+
{
|
| 1757 |
+
"epoch": 129.93,
|
| 1758 |
+
"eval_loss": 0.16691875457763672,
|
| 1759 |
+
"eval_runtime": 118.5844,
|
| 1760 |
+
"eval_samples_per_second": 107.552,
|
| 1761 |
+
"eval_steps_per_second": 3.365,
|
| 1762 |
+
"eval_wer": 0.15731911620605002,
|
| 1763 |
+
"step": 234000
|
| 1764 |
+
},
|
| 1765 |
+
{
|
| 1766 |
+
"epoch": 131.04,
|
| 1767 |
+
"learning_rate": 5.6368333333333334e-05,
|
| 1768 |
+
"loss": 0.4335,
|
| 1769 |
+
"step": 236000
|
| 1770 |
+
},
|
| 1771 |
+
{
|
| 1772 |
+
"epoch": 131.04,
|
| 1773 |
+
"eval_loss": 0.17210978269577026,
|
| 1774 |
+
"eval_runtime": 110.1221,
|
| 1775 |
+
"eval_samples_per_second": 115.817,
|
| 1776 |
+
"eval_steps_per_second": 3.623,
|
| 1777 |
+
"eval_wer": 0.15840436628770718,
|
| 1778 |
+
"step": 236000
|
| 1779 |
+
},
|
| 1780 |
+
{
|
| 1781 |
+
"epoch": 132.15,
|
| 1782 |
+
"learning_rate": 5.599796296296297e-05,
|
| 1783 |
+
"loss": 0.4302,
|
| 1784 |
+
"step": 238000
|
| 1785 |
+
},
|
| 1786 |
+
{
|
| 1787 |
+
"epoch": 132.15,
|
| 1788 |
+
"eval_loss": 0.17578274011611938,
|
| 1789 |
+
"eval_runtime": 110.8686,
|
| 1790 |
+
"eval_samples_per_second": 115.037,
|
| 1791 |
+
"eval_steps_per_second": 3.599,
|
| 1792 |
+
"eval_wer": 0.1587309949530603,
|
| 1793 |
+
"step": 238000
|
| 1794 |
+
},
|
| 1795 |
+
{
|
| 1796 |
+
"epoch": 133.26,
|
| 1797 |
+
"learning_rate": 5.562777777777778e-05,
|
| 1798 |
+
"loss": 0.4292,
|
| 1799 |
+
"step": 240000
|
| 1800 |
+
},
|
| 1801 |
+
{
|
| 1802 |
+
"epoch": 133.26,
|
| 1803 |
+
"eval_loss": 0.173477903008461,
|
| 1804 |
+
"eval_runtime": 110.0323,
|
| 1805 |
+
"eval_samples_per_second": 115.911,
|
| 1806 |
+
"eval_steps_per_second": 3.626,
|
| 1807 |
+
"eval_wer": 0.16007965524871193,
|
| 1808 |
+
"step": 240000
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 134.37,
|
| 1812 |
+
"learning_rate": 5.5257407407407414e-05,
|
| 1813 |
+
"loss": 0.4269,
|
| 1814 |
+
"step": 242000
|
| 1815 |
+
},
|
| 1816 |
+
{
|
| 1817 |
+
"epoch": 134.37,
|
| 1818 |
+
"eval_loss": 0.16973139345645905,
|
| 1819 |
+
"eval_runtime": 113.6422,
|
| 1820 |
+
"eval_samples_per_second": 112.229,
|
| 1821 |
+
"eval_steps_per_second": 3.511,
|
| 1822 |
+
"eval_wer": 0.1569819511321371,
|
| 1823 |
+
"step": 242000
|
| 1824 |
+
},
|
| 1825 |
+
{
|
| 1826 |
+
"epoch": 135.48,
|
| 1827 |
+
"learning_rate": 5.488722222222222e-05,
|
| 1828 |
+
"loss": 0.4273,
|
| 1829 |
+
"step": 244000
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 135.48,
|
| 1833 |
+
"eval_loss": 0.17380733788013458,
|
| 1834 |
+
"eval_runtime": 110.4957,
|
| 1835 |
+
"eval_samples_per_second": 115.425,
|
| 1836 |
+
"eval_steps_per_second": 3.611,
|
| 1837 |
+
"eval_wer": 0.15647620352126773,
|
| 1838 |
+
"step": 244000
|
| 1839 |
+
},
|
| 1840 |
+
{
|
| 1841 |
+
"epoch": 136.59,
|
| 1842 |
+
"learning_rate": 5.4517037037037045e-05,
|
| 1843 |
+
"loss": 0.4257,
|
| 1844 |
+
"step": 246000
|
| 1845 |
+
},
|
| 1846 |
+
{
|
| 1847 |
+
"epoch": 136.59,
|
| 1848 |
+
"eval_loss": 0.16813333332538605,
|
| 1849 |
+
"eval_runtime": 111.9779,
|
| 1850 |
+
"eval_samples_per_second": 113.897,
|
| 1851 |
+
"eval_steps_per_second": 3.563,
|
| 1852 |
+
"eval_wer": 0.15652888556406663,
|
| 1853 |
+
"step": 246000
|
| 1854 |
+
},
|
| 1855 |
+
{
|
| 1856 |
+
"epoch": 137.7,
|
| 1857 |
+
"learning_rate": 5.414666666666667e-05,
|
| 1858 |
+
"loss": 0.4221,
|
| 1859 |
+
"step": 248000
|
| 1860 |
+
},
|
| 1861 |
+
{
|
| 1862 |
+
"epoch": 137.7,
|
| 1863 |
+
"eval_loss": 0.17320315539836884,
|
| 1864 |
+
"eval_runtime": 111.5731,
|
| 1865 |
+
"eval_samples_per_second": 114.311,
|
| 1866 |
+
"eval_steps_per_second": 3.576,
|
| 1867 |
+
"eval_wer": 0.15928942460672854,
|
| 1868 |
+
"step": 248000
|
| 1869 |
+
},
|
| 1870 |
+
{
|
| 1871 |
+
"epoch": 138.81,
|
| 1872 |
+
"learning_rate": 5.377648148148149e-05,
|
| 1873 |
+
"loss": 0.422,
|
| 1874 |
+
"step": 250000
|
| 1875 |
+
},
|
| 1876 |
+
{
|
| 1877 |
+
"epoch": 138.81,
|
| 1878 |
+
"eval_loss": 0.17005948722362518,
|
| 1879 |
+
"eval_runtime": 116.6285,
|
| 1880 |
+
"eval_samples_per_second": 109.356,
|
| 1881 |
+
"eval_steps_per_second": 3.421,
|
| 1882 |
+
"eval_wer": 0.15444267666923053,
|
| 1883 |
+
"step": 250000
|
| 1884 |
+
},
|
| 1885 |
+
{
|
| 1886 |
+
"epoch": 139.92,
|
| 1887 |
+
"learning_rate": 5.34062962962963e-05,
|
| 1888 |
+
"loss": 0.4178,
|
| 1889 |
+
"step": 252000
|
| 1890 |
+
},
|
| 1891 |
+
{
|
| 1892 |
+
"epoch": 139.92,
|
| 1893 |
+
"eval_loss": 0.17322979867458344,
|
| 1894 |
+
"eval_runtime": 117.6422,
|
| 1895 |
+
"eval_samples_per_second": 108.413,
|
| 1896 |
+
"eval_steps_per_second": 3.392,
|
| 1897 |
+
"eval_wer": 0.15422141208947518,
|
| 1898 |
+
"step": 252000
|
| 1899 |
+
},
|
| 1900 |
+
{
|
| 1901 |
+
"epoch": 141.03,
|
| 1902 |
+
"learning_rate": 5.303611111111111e-05,
|
| 1903 |
+
"loss": 0.4182,
|
| 1904 |
+
"step": 254000
|
| 1905 |
+
},
|
| 1906 |
+
{
|
| 1907 |
+
"epoch": 141.03,
|
| 1908 |
+
"eval_loss": 0.1738407015800476,
|
| 1909 |
+
"eval_runtime": 113.7677,
|
| 1910 |
+
"eval_samples_per_second": 112.106,
|
| 1911 |
+
"eval_steps_per_second": 3.507,
|
| 1912 |
+
"eval_wer": 0.15336796299613314,
|
| 1913 |
+
"step": 254000
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 142.14,
|
| 1917 |
+
"learning_rate": 5.266592592592593e-05,
|
| 1918 |
+
"loss": 0.4166,
|
| 1919 |
+
"step": 256000
|
| 1920 |
+
},
|
| 1921 |
+
{
|
| 1922 |
+
"epoch": 142.14,
|
| 1923 |
+
"eval_loss": 0.1734342724084854,
|
| 1924 |
+
"eval_runtime": 112.0095,
|
| 1925 |
+
"eval_samples_per_second": 113.865,
|
| 1926 |
+
"eval_steps_per_second": 3.562,
|
| 1927 |
+
"eval_wer": 0.15277792411678556,
|
| 1928 |
+
"step": 256000
|
| 1929 |
+
},
|
| 1930 |
+
{
|
| 1931 |
+
"epoch": 143.25,
|
| 1932 |
+
"learning_rate": 5.229574074074074e-05,
|
| 1933 |
+
"loss": 0.4163,
|
| 1934 |
+
"step": 258000
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 143.25,
|
| 1938 |
+
"eval_loss": 0.1752772331237793,
|
| 1939 |
+
"eval_runtime": 111.2622,
|
| 1940 |
+
"eval_samples_per_second": 114.63,
|
| 1941 |
+
"eval_steps_per_second": 3.586,
|
| 1942 |
+
"eval_wer": 0.15545417189096924,
|
| 1943 |
+
"step": 258000
|
| 1944 |
+
},
|
| 1945 |
+
{
|
| 1946 |
+
"epoch": 144.36,
|
| 1947 |
+
"learning_rate": 5.1925370370370374e-05,
|
| 1948 |
+
"loss": 0.4144,
|
| 1949 |
+
"step": 260000
|
| 1950 |
+
},
|
| 1951 |
+
{
|
| 1952 |
+
"epoch": 144.36,
|
| 1953 |
+
"eval_loss": 0.17757321894168854,
|
| 1954 |
+
"eval_runtime": 114.1149,
|
| 1955 |
+
"eval_samples_per_second": 111.765,
|
| 1956 |
+
"eval_steps_per_second": 3.496,
|
| 1957 |
+
"eval_wer": 0.15680283218662086,
|
| 1958 |
+
"step": 260000
|
| 1959 |
+
},
|
| 1960 |
+
{
|
| 1961 |
+
"epoch": 145.47,
|
| 1962 |
+
"learning_rate": 5.1555e-05,
|
| 1963 |
+
"loss": 0.4103,
|
| 1964 |
+
"step": 262000
|
| 1965 |
+
},
|
| 1966 |
+
{
|
| 1967 |
+
"epoch": 145.47,
|
| 1968 |
+
"eval_loss": 0.17166772484779358,
|
| 1969 |
+
"eval_runtime": 111.5777,
|
| 1970 |
+
"eval_samples_per_second": 114.306,
|
| 1971 |
+
"eval_steps_per_second": 3.576,
|
| 1972 |
+
"eval_wer": 0.15237754059151398,
|
| 1973 |
+
"step": 262000
|
| 1974 |
+
},
|
| 1975 |
+
{
|
| 1976 |
+
"epoch": 146.59,
|
| 1977 |
+
"learning_rate": 5.1185000000000005e-05,
|
| 1978 |
+
"loss": 0.4117,
|
| 1979 |
+
"step": 264000
|
| 1980 |
+
},
|
| 1981 |
+
{
|
| 1982 |
+
"epoch": 146.59,
|
| 1983 |
+
"eval_loss": 0.1743527501821518,
|
| 1984 |
+
"eval_runtime": 112.423,
|
| 1985 |
+
"eval_samples_per_second": 113.447,
|
| 1986 |
+
"eval_steps_per_second": 3.549,
|
| 1987 |
+
"eval_wer": 0.15553846315944747,
|
| 1988 |
+
"step": 264000
|
| 1989 |
+
},
|
| 1990 |
+
{
|
| 1991 |
+
"epoch": 147.7,
|
| 1992 |
+
"learning_rate": 5.081462962962963e-05,
|
| 1993 |
+
"loss": 0.4085,
|
| 1994 |
+
"step": 266000
|
| 1995 |
+
},
|
| 1996 |
+
{
|
| 1997 |
+
"epoch": 147.7,
|
| 1998 |
+
"eval_loss": 0.17306753993034363,
|
| 1999 |
+
"eval_runtime": 112.3063,
|
| 2000 |
+
"eval_samples_per_second": 113.564,
|
| 2001 |
+
"eval_steps_per_second": 3.553,
|
| 2002 |
+
"eval_wer": 0.15564382724504525,
|
| 2003 |
+
"step": 266000
|
| 2004 |
+
},
|
| 2005 |
+
{
|
| 2006 |
+
"epoch": 148.81,
|
| 2007 |
+
"learning_rate": 5.0444259259259264e-05,
|
| 2008 |
+
"loss": 0.408,
|
| 2009 |
+
"step": 268000
|
| 2010 |
+
},
|
| 2011 |
+
{
|
| 2012 |
+
"epoch": 148.81,
|
| 2013 |
+
"eval_loss": 0.17164543271064758,
|
| 2014 |
+
"eval_runtime": 113.1307,
|
| 2015 |
+
"eval_samples_per_second": 112.737,
|
| 2016 |
+
"eval_steps_per_second": 3.527,
|
| 2017 |
+
"eval_wer": 0.1530940163735789,
|
| 2018 |
+
"step": 268000
|
| 2019 |
+
},
|
| 2020 |
+
{
|
| 2021 |
+
"epoch": 149.92,
|
| 2022 |
+
"learning_rate": 5.0073888888888886e-05,
|
| 2023 |
+
"loss": 0.4078,
|
| 2024 |
+
"step": 270000
|
| 2025 |
+
},
|
| 2026 |
+
{
|
| 2027 |
+
"epoch": 149.92,
|
| 2028 |
+
"eval_loss": 0.17812077701091766,
|
| 2029 |
+
"eval_runtime": 114.4521,
|
| 2030 |
+
"eval_samples_per_second": 111.435,
|
| 2031 |
+
"eval_steps_per_second": 3.486,
|
| 2032 |
+
"eval_wer": 0.15219842164599776,
|
| 2033 |
+
"step": 270000
|
| 2034 |
+
},
|
| 2035 |
+
{
|
| 2036 |
+
"epoch": 151.03,
|
| 2037 |
+
"learning_rate": 4.970425925925926e-05,
|
| 2038 |
+
"loss": 0.4051,
|
| 2039 |
+
"step": 272000
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"epoch": 151.03,
|
| 2043 |
+
"eval_loss": 0.17345324158668518,
|
| 2044 |
+
"eval_runtime": 109.0896,
|
| 2045 |
+
"eval_samples_per_second": 116.913,
|
| 2046 |
+
"eval_steps_per_second": 3.658,
|
| 2047 |
+
"eval_wer": 0.15237754059151398,
|
| 2048 |
+
"step": 272000
|
| 2049 |
+
},
|
| 2050 |
+
{
|
| 2051 |
+
"epoch": 152.14,
|
| 2052 |
+
"learning_rate": 4.933388888888889e-05,
|
| 2053 |
+
"loss": 0.4033,
|
| 2054 |
+
"step": 274000
|
| 2055 |
+
},
|
| 2056 |
+
{
|
| 2057 |
+
"epoch": 152.14,
|
| 2058 |
+
"eval_loss": 0.18191710114479065,
|
| 2059 |
+
"eval_runtime": 111.0027,
|
| 2060 |
+
"eval_samples_per_second": 114.898,
|
| 2061 |
+
"eval_steps_per_second": 3.595,
|
| 2062 |
+
"eval_wer": 0.15185072016352505,
|
| 2063 |
+
"step": 274000
|
| 2064 |
+
},
|
| 2065 |
+
{
|
| 2066 |
+
"epoch": 153.25,
|
| 2067 |
+
"learning_rate": 4.896351851851852e-05,
|
| 2068 |
+
"loss": 0.4037,
|
| 2069 |
+
"step": 276000
|
| 2070 |
+
},
|
| 2071 |
+
{
|
| 2072 |
+
"epoch": 153.25,
|
| 2073 |
+
"eval_loss": 0.18181933462619781,
|
| 2074 |
+
"eval_runtime": 110.4212,
|
| 2075 |
+
"eval_samples_per_second": 115.503,
|
| 2076 |
+
"eval_steps_per_second": 3.613,
|
| 2077 |
+
"eval_wer": 0.15307294355645934,
|
| 2078 |
+
"step": 276000
|
| 2079 |
+
},
|
| 2080 |
+
{
|
| 2081 |
+
"epoch": 154.36,
|
| 2082 |
+
"learning_rate": 4.8593333333333335e-05,
|
| 2083 |
+
"loss": 0.4002,
|
| 2084 |
+
"step": 278000
|
| 2085 |
+
},
|
| 2086 |
+
{
|
| 2087 |
+
"epoch": 154.36,
|
| 2088 |
+
"eval_loss": 0.17859378457069397,
|
| 2089 |
+
"eval_runtime": 116.538,
|
| 2090 |
+
"eval_samples_per_second": 109.441,
|
| 2091 |
+
"eval_steps_per_second": 3.424,
|
| 2092 |
+
"eval_wer": 0.1548852058287412,
|
| 2093 |
+
"step": 278000
|
| 2094 |
+
},
|
| 2095 |
+
{
|
| 2096 |
+
"epoch": 155.47,
|
| 2097 |
+
"learning_rate": 4.8222962962962964e-05,
|
| 2098 |
+
"loss": 0.4021,
|
| 2099 |
+
"step": 280000
|
| 2100 |
+
},
|
| 2101 |
+
{
|
| 2102 |
+
"epoch": 155.47,
|
| 2103 |
+
"eval_loss": 0.17952340841293335,
|
| 2104 |
+
"eval_runtime": 112.6278,
|
| 2105 |
+
"eval_samples_per_second": 113.24,
|
| 2106 |
+
"eval_steps_per_second": 3.543,
|
| 2107 |
+
"eval_wer": 0.15223003087167708,
|
| 2108 |
+
"step": 280000
|
| 2109 |
+
},
|
| 2110 |
+
{
|
| 2111 |
+
"epoch": 156.58,
|
| 2112 |
+
"learning_rate": 4.785277777777778e-05,
|
| 2113 |
+
"loss": 0.3985,
|
| 2114 |
+
"step": 282000
|
| 2115 |
+
},
|
| 2116 |
+
{
|
| 2117 |
+
"epoch": 156.58,
|
| 2118 |
+
"eval_loss": 0.18241500854492188,
|
| 2119 |
+
"eval_runtime": 110.8984,
|
| 2120 |
+
"eval_samples_per_second": 115.006,
|
| 2121 |
+
"eval_steps_per_second": 3.598,
|
| 2122 |
+
"eval_wer": 0.1529570430623018,
|
| 2123 |
+
"step": 282000
|
| 2124 |
+
},
|
| 2125 |
+
{
|
| 2126 |
+
"epoch": 157.69,
|
| 2127 |
+
"learning_rate": 4.7482592592592595e-05,
|
| 2128 |
+
"loss": 0.3987,
|
| 2129 |
+
"step": 284000
|
| 2130 |
+
},
|
| 2131 |
+
{
|
| 2132 |
+
"epoch": 157.69,
|
| 2133 |
+
"eval_loss": 0.18239329755306244,
|
| 2134 |
+
"eval_runtime": 116.2871,
|
| 2135 |
+
"eval_samples_per_second": 109.677,
|
| 2136 |
+
"eval_steps_per_second": 3.431,
|
| 2137 |
+
"eval_wer": 0.15229324932303576,
|
| 2138 |
+
"step": 284000
|
| 2139 |
+
},
|
| 2140 |
+
{
|
| 2141 |
+
"epoch": 158.8,
|
| 2142 |
+
"learning_rate": 4.7112222222222224e-05,
|
| 2143 |
+
"loss": 0.3957,
|
| 2144 |
+
"step": 286000
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 158.8,
|
| 2148 |
+
"eval_loss": 0.18385788798332214,
|
| 2149 |
+
"eval_runtime": 111.0629,
|
| 2150 |
+
"eval_samples_per_second": 114.836,
|
| 2151 |
+
"eval_steps_per_second": 3.593,
|
| 2152 |
+
"eval_wer": 0.15160838276665015,
|
| 2153 |
+
"step": 286000
|
| 2154 |
+
}
|
| 2155 |
+
],
|
| 2156 |
+
"max_steps": 540300,
|
| 2157 |
+
"num_train_epochs": 300,
|
| 2158 |
+
"total_flos": 2.3864414600426045e+21,
|
| 2159 |
+
"trial_name": null,
|
| 2160 |
+
"trial_params": null
|
| 2161 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c80bf88c5219e3a6a031434b7830428c5cef13c7fb2a42694ca992104136c0b6
|
| 3 |
+
size 2991
|
vocab.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"\\": 1, "]": 2, "`": 3, "a": 4, "b": 5, "c": 6, "d": 7, "e": 8, "f": 9, "g": 10, "h": 11, "i": 12, "j": 13, "k": 14, "l": 15, "m": 16, "n": 17, "o": 18, "p": 19, "q": 20, "r": 21, "s": 22, "t": 23, "u": 24, "v": 25, "w": 26, "x": 27, "y": 28, "z": 29, "½": 30, "á": 31, "é": 32, "ó": 33, "ŋ": 34, "š": 35, "η": 36, "ղ": 37, "῎": 38, "῝": 39, "`": 40, "–": 41, "—": 42, "“": 43, "”": 44, "‟": 45, "•": 46, "…": 47, "|": 0, "[UNK]": 48, "[PAD]": 49}
|