Upload 8 files
Browse files- config.json +82 -0
- model.safetensors +3 -0
- preprocessor_config.json +10 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- trainer_state.json +1883 -0
- training_args.bin +3 -0
- vocab.json +1 -0
config.json
ADDED
|
@@ -0,0 +1,82 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "facebook/w2v-bert-2.0",
|
| 3 |
+
"activation_dropout": 0.0,
|
| 4 |
+
"adapter_act": "relu",
|
| 5 |
+
"adapter_kernel_size": 3,
|
| 6 |
+
"adapter_stride": 2,
|
| 7 |
+
"add_adapter": true,
|
| 8 |
+
"apply_spec_augment": false,
|
| 9 |
+
"architectures": [
|
| 10 |
+
"Wav2Vec2BertForCTC"
|
| 11 |
+
],
|
| 12 |
+
"attention_dropout": 0.2,
|
| 13 |
+
"bos_token_id": 1,
|
| 14 |
+
"classifier_proj_size": 768,
|
| 15 |
+
"codevector_dim": 768,
|
| 16 |
+
"conformer_conv_dropout": 0.1,
|
| 17 |
+
"contrastive_logits_temperature": 0.1,
|
| 18 |
+
"conv_depthwise_kernel_size": 31,
|
| 19 |
+
"ctc_loss_reduction": "mean",
|
| 20 |
+
"ctc_zero_infinity": true,
|
| 21 |
+
"diversity_loss_weight": 0.1,
|
| 22 |
+
"eos_token_id": 2,
|
| 23 |
+
"feat_proj_dropout": 0.2,
|
| 24 |
+
"feat_quantizer_dropout": 0.0,
|
| 25 |
+
"feature_projection_input_dim": 160,
|
| 26 |
+
"final_dropout": 0.1,
|
| 27 |
+
"hidden_act": "swish",
|
| 28 |
+
"hidden_dropout": 0.2,
|
| 29 |
+
"hidden_size": 1024,
|
| 30 |
+
"initializer_range": 0.02,
|
| 31 |
+
"intermediate_size": 4096,
|
| 32 |
+
"layer_norm_eps": 1e-05,
|
| 33 |
+
"layerdrop": 0.2,
|
| 34 |
+
"left_max_position_embeddings": 64,
|
| 35 |
+
"mask_feature_length": 10,
|
| 36 |
+
"mask_feature_min_masks": 0,
|
| 37 |
+
"mask_feature_prob": 0.0,
|
| 38 |
+
"mask_time_length": 10,
|
| 39 |
+
"mask_time_min_masks": 2,
|
| 40 |
+
"mask_time_prob": 0.0,
|
| 41 |
+
"max_source_positions": 5000,
|
| 42 |
+
"model_type": "wav2vec2-bert",
|
| 43 |
+
"num_adapter_layers": 1,
|
| 44 |
+
"num_attention_heads": 16,
|
| 45 |
+
"num_codevector_groups": 2,
|
| 46 |
+
"num_codevectors_per_group": 320,
|
| 47 |
+
"num_hidden_layers": 24,
|
| 48 |
+
"num_negatives": 100,
|
| 49 |
+
"output_hidden_size": 1024,
|
| 50 |
+
"pad_token_id": 64,
|
| 51 |
+
"position_embeddings_type": "relative_key",
|
| 52 |
+
"proj_codevector_dim": 768,
|
| 53 |
+
"right_max_position_embeddings": 8,
|
| 54 |
+
"rotary_embedding_base": 10000,
|
| 55 |
+
"tdnn_dilation": [
|
| 56 |
+
1,
|
| 57 |
+
2,
|
| 58 |
+
3,
|
| 59 |
+
1,
|
| 60 |
+
1
|
| 61 |
+
],
|
| 62 |
+
"tdnn_dim": [
|
| 63 |
+
512,
|
| 64 |
+
512,
|
| 65 |
+
512,
|
| 66 |
+
512,
|
| 67 |
+
1500
|
| 68 |
+
],
|
| 69 |
+
"tdnn_kernel": [
|
| 70 |
+
5,
|
| 71 |
+
3,
|
| 72 |
+
3,
|
| 73 |
+
1,
|
| 74 |
+
1
|
| 75 |
+
],
|
| 76 |
+
"torch_dtype": "float32",
|
| 77 |
+
"transformers_version": "4.42.4",
|
| 78 |
+
"use_intermediate_ffn_before_adapter": false,
|
| 79 |
+
"use_weighted_layer_sum": false,
|
| 80 |
+
"vocab_size": 67,
|
| 81 |
+
"xvector_output_dim": 512
|
| 82 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:04d1af1948f55a4eaefcd56200d21216520cea00842fc2e5a6f32c953e1add50
|
| 3 |
+
size 2423089260
|
preprocessor_config.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"feature_extractor_type": "SeamlessM4TFeatureExtractor",
|
| 3 |
+
"feature_size": 80,
|
| 4 |
+
"num_mel_bins": 80,
|
| 5 |
+
"padding_side": "right",
|
| 6 |
+
"padding_value": 0.0,
|
| 7 |
+
"return_attention_mask": true,
|
| 8 |
+
"sampling_rate": 16000,
|
| 9 |
+
"stride": 2
|
| 10 |
+
}
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:40c2efbd34a555c011bddc6c43f76a836326617ad12bb332da4cf7a1afc0c347
|
| 3 |
+
size 14244
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a963b078720dd9db782f52ebc335fcd45d25e21fd109960da0e1be793c5a94fc
|
| 3 |
+
size 1256
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1883 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.35703112038112395,
|
| 3 |
+
"best_model_checkpoint": "results/punjabi_model/20240829_200444/checkpoint-3000",
|
| 4 |
+
"epoch": 9.999656616990592,
|
| 5 |
+
"eval_steps": 3000,
|
| 6 |
+
"global_step": 109200,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.045784401254492596,
|
| 13 |
+
"grad_norm": 2.4544355869293213,
|
| 14 |
+
"learning_rate": 9.980000000000001e-06,
|
| 15 |
+
"loss": 3.9787,
|
| 16 |
+
"step": 500
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.09156880250898519,
|
| 20 |
+
"grad_norm": 5.300338268280029,
|
| 21 |
+
"learning_rate": 1.9980000000000002e-05,
|
| 22 |
+
"loss": 2.7973,
|
| 23 |
+
"step": 1000
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.1373532037634778,
|
| 27 |
+
"grad_norm": 5.187418460845947,
|
| 28 |
+
"learning_rate": 2.9980000000000004e-05,
|
| 29 |
+
"loss": 2.1081,
|
| 30 |
+
"step": 1500
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.18313760501797038,
|
| 34 |
+
"grad_norm": 5.451952934265137,
|
| 35 |
+
"learning_rate": 3.998000000000001e-05,
|
| 36 |
+
"loss": 1.7718,
|
| 37 |
+
"step": 2000
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.22892200627246298,
|
| 41 |
+
"grad_norm": 10.363802909851074,
|
| 42 |
+
"learning_rate": 3.962053231939164e-05,
|
| 43 |
+
"loss": 1.6058,
|
| 44 |
+
"step": 2500
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.2747064075269556,
|
| 48 |
+
"grad_norm": 3.4008052349090576,
|
| 49 |
+
"learning_rate": 3.924030418250951e-05,
|
| 50 |
+
"loss": 1.5005,
|
| 51 |
+
"step": 3000
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.2747064075269556,
|
| 55 |
+
"eval_loss": 0.580813467502594,
|
| 56 |
+
"eval_runtime": 645.4536,
|
| 57 |
+
"eval_samples_per_second": 10.151,
|
| 58 |
+
"eval_steps_per_second": 10.151,
|
| 59 |
+
"eval_wer": 0.35703112038112395,
|
| 60 |
+
"step": 3000
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 0.32049080878144814,
|
| 64 |
+
"grad_norm": 2.9593212604522705,
|
| 65 |
+
"learning_rate": 3.886007604562738e-05,
|
| 66 |
+
"loss": 1.4034,
|
| 67 |
+
"step": 3500
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 0.36627521003594077,
|
| 71 |
+
"grad_norm": 4.573808193206787,
|
| 72 |
+
"learning_rate": 3.847984790874525e-05,
|
| 73 |
+
"loss": 1.3577,
|
| 74 |
+
"step": 4000
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.41205961129043334,
|
| 78 |
+
"grad_norm": 2.684593915939331,
|
| 79 |
+
"learning_rate": 3.809961977186312e-05,
|
| 80 |
+
"loss": 1.3114,
|
| 81 |
+
"step": 4500
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 0.45784401254492596,
|
| 85 |
+
"grad_norm": 2.6589484214782715,
|
| 86 |
+
"learning_rate": 3.771939163498099e-05,
|
| 87 |
+
"loss": 1.3163,
|
| 88 |
+
"step": 5000
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 0.5036284137994186,
|
| 92 |
+
"grad_norm": 3.144775390625,
|
| 93 |
+
"learning_rate": 3.733992395437262e-05,
|
| 94 |
+
"loss": 1.2375,
|
| 95 |
+
"step": 5500
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 0.5494128150539112,
|
| 99 |
+
"grad_norm": 3.1532704830169678,
|
| 100 |
+
"learning_rate": 3.6959695817490496e-05,
|
| 101 |
+
"loss": 1.2442,
|
| 102 |
+
"step": 6000
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 0.5494128150539112,
|
| 106 |
+
"eval_loss": 0.43034785985946655,
|
| 107 |
+
"eval_runtime": 665.7368,
|
| 108 |
+
"eval_samples_per_second": 9.842,
|
| 109 |
+
"eval_steps_per_second": 9.842,
|
| 110 |
+
"eval_wer": 0.2554395487189211,
|
| 111 |
+
"step": 6000
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 0.5951972163084037,
|
| 115 |
+
"grad_norm": 9.455696105957031,
|
| 116 |
+
"learning_rate": 3.657946768060837e-05,
|
| 117 |
+
"loss": 1.225,
|
| 118 |
+
"step": 6500
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 0.6409816175628963,
|
| 122 |
+
"grad_norm": 3.3579087257385254,
|
| 123 |
+
"learning_rate": 3.619923954372624e-05,
|
| 124 |
+
"loss": 1.1457,
|
| 125 |
+
"step": 7000
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 0.686766018817389,
|
| 129 |
+
"grad_norm": 4.421695709228516,
|
| 130 |
+
"learning_rate": 3.581901140684411e-05,
|
| 131 |
+
"loss": 1.1656,
|
| 132 |
+
"step": 7500
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 0.7325504200718815,
|
| 136 |
+
"grad_norm": 3.5701684951782227,
|
| 137 |
+
"learning_rate": 3.5439543726235745e-05,
|
| 138 |
+
"loss": 1.175,
|
| 139 |
+
"step": 8000
|
| 140 |
+
},
|
| 141 |
+
{
|
| 142 |
+
"epoch": 0.7783348213263741,
|
| 143 |
+
"grad_norm": 4.08390474319458,
|
| 144 |
+
"learning_rate": 3.505931558935361e-05,
|
| 145 |
+
"loss": 1.131,
|
| 146 |
+
"step": 8500
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 0.8241192225808667,
|
| 150 |
+
"grad_norm": 3.4614651203155518,
|
| 151 |
+
"learning_rate": 3.4679847908745255e-05,
|
| 152 |
+
"loss": 1.1133,
|
| 153 |
+
"step": 9000
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 0.8241192225808667,
|
| 157 |
+
"eval_loss": 0.3901245594024658,
|
| 158 |
+
"eval_runtime": 664.9166,
|
| 159 |
+
"eval_samples_per_second": 9.854,
|
| 160 |
+
"eval_steps_per_second": 9.854,
|
| 161 |
+
"eval_wer": 0.2110935981607452,
|
| 162 |
+
"step": 9000
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 0.8699036238353592,
|
| 166 |
+
"grad_norm": 2.597609519958496,
|
| 167 |
+
"learning_rate": 3.429961977186312e-05,
|
| 168 |
+
"loss": 1.0977,
|
| 169 |
+
"step": 9500
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 0.9156880250898519,
|
| 173 |
+
"grad_norm": 3.4706475734710693,
|
| 174 |
+
"learning_rate": 3.3919391634980995e-05,
|
| 175 |
+
"loss": 1.0986,
|
| 176 |
+
"step": 10000
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 0.9614724263443445,
|
| 180 |
+
"grad_norm": 2.1491544246673584,
|
| 181 |
+
"learning_rate": 3.353916349809886e-05,
|
| 182 |
+
"loss": 1.1154,
|
| 183 |
+
"step": 10500
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 1.0072568275988372,
|
| 187 |
+
"grad_norm": 1.495735764503479,
|
| 188 |
+
"learning_rate": 3.3158935361216734e-05,
|
| 189 |
+
"loss": 1.1157,
|
| 190 |
+
"step": 11000
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 1.0530412288533297,
|
| 194 |
+
"grad_norm": 2.120736837387085,
|
| 195 |
+
"learning_rate": 3.277870722433461e-05,
|
| 196 |
+
"loss": 1.0495,
|
| 197 |
+
"step": 11500
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 1.0988256301078223,
|
| 201 |
+
"grad_norm": 2.2971479892730713,
|
| 202 |
+
"learning_rate": 3.239847908745247e-05,
|
| 203 |
+
"loss": 1.0785,
|
| 204 |
+
"step": 12000
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 1.0988256301078223,
|
| 208 |
+
"eval_loss": 0.34569498896598816,
|
| 209 |
+
"eval_runtime": 655.7186,
|
| 210 |
+
"eval_samples_per_second": 9.992,
|
| 211 |
+
"eval_steps_per_second": 9.992,
|
| 212 |
+
"eval_wer": 0.19471569766063854,
|
| 213 |
+
"step": 12000
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 1.1446100313623149,
|
| 217 |
+
"grad_norm": 1.080255150794983,
|
| 218 |
+
"learning_rate": 3.2018250950570346e-05,
|
| 219 |
+
"loss": 1.0311,
|
| 220 |
+
"step": 12500
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 1.1903944326168074,
|
| 224 |
+
"grad_norm": 2.3899152278900146,
|
| 225 |
+
"learning_rate": 3.163802281368821e-05,
|
| 226 |
+
"loss": 1.107,
|
| 227 |
+
"step": 13000
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 1.2361788338713,
|
| 231 |
+
"grad_norm": 2.9057199954986572,
|
| 232 |
+
"learning_rate": 3.125855513307985e-05,
|
| 233 |
+
"loss": 1.0711,
|
| 234 |
+
"step": 13500
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 1.2819632351257926,
|
| 238 |
+
"grad_norm": 1.879520058631897,
|
| 239 |
+
"learning_rate": 3.087832699619772e-05,
|
| 240 |
+
"loss": 1.0348,
|
| 241 |
+
"step": 14000
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 1.3277476363802854,
|
| 245 |
+
"grad_norm": 2.777270555496216,
|
| 246 |
+
"learning_rate": 3.0498098859315592e-05,
|
| 247 |
+
"loss": 1.0308,
|
| 248 |
+
"step": 14500
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 1.373532037634778,
|
| 252 |
+
"grad_norm": 3.571882724761963,
|
| 253 |
+
"learning_rate": 3.0117870722433462e-05,
|
| 254 |
+
"loss": 1.0245,
|
| 255 |
+
"step": 15000
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 1.373532037634778,
|
| 259 |
+
"eval_loss": 0.3307068943977356,
|
| 260 |
+
"eval_runtime": 655.2961,
|
| 261 |
+
"eval_samples_per_second": 9.999,
|
| 262 |
+
"eval_steps_per_second": 9.999,
|
| 263 |
+
"eval_wer": 0.1764061055675381,
|
| 264 |
+
"step": 15000
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 1.4193164388892705,
|
| 268 |
+
"grad_norm": 2.154317855834961,
|
| 269 |
+
"learning_rate": 2.973764258555133e-05,
|
| 270 |
+
"loss": 1.0197,
|
| 271 |
+
"step": 15500
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 1.465100840143763,
|
| 275 |
+
"grad_norm": 2.7176313400268555,
|
| 276 |
+
"learning_rate": 2.9357414448669205e-05,
|
| 277 |
+
"loss": 1.0042,
|
| 278 |
+
"step": 16000
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 1.5108852413982556,
|
| 282 |
+
"grad_norm": 3.9277138710021973,
|
| 283 |
+
"learning_rate": 2.8977946768060842e-05,
|
| 284 |
+
"loss": 0.9801,
|
| 285 |
+
"step": 16500
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 1.5566696426527482,
|
| 289 |
+
"grad_norm": 6.810076713562012,
|
| 290 |
+
"learning_rate": 2.859771863117871e-05,
|
| 291 |
+
"loss": 0.9256,
|
| 292 |
+
"step": 17000
|
| 293 |
+
},
|
| 294 |
+
{
|
| 295 |
+
"epoch": 1.6024540439072408,
|
| 296 |
+
"grad_norm": 1.7569458484649658,
|
| 297 |
+
"learning_rate": 2.821749049429658e-05,
|
| 298 |
+
"loss": 0.981,
|
| 299 |
+
"step": 17500
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 1.6482384451617333,
|
| 303 |
+
"grad_norm": 3.3207457065582275,
|
| 304 |
+
"learning_rate": 2.783726235741445e-05,
|
| 305 |
+
"loss": 0.9452,
|
| 306 |
+
"step": 18000
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 1.6482384451617333,
|
| 310 |
+
"eval_loss": 0.3014201521873474,
|
| 311 |
+
"eval_runtime": 651.2443,
|
| 312 |
+
"eval_samples_per_second": 10.061,
|
| 313 |
+
"eval_steps_per_second": 10.061,
|
| 314 |
+
"eval_wer": 0.1733545068853547,
|
| 315 |
+
"step": 18000
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 1.694022846416226,
|
| 319 |
+
"grad_norm": 1.4468517303466797,
|
| 320 |
+
"learning_rate": 2.7457034220532324e-05,
|
| 321 |
+
"loss": 0.9969,
|
| 322 |
+
"step": 18500
|
| 323 |
+
},
|
| 324 |
+
{
|
| 325 |
+
"epoch": 1.7398072476707185,
|
| 326 |
+
"grad_norm": 1.2308686971664429,
|
| 327 |
+
"learning_rate": 2.7077566539923958e-05,
|
| 328 |
+
"loss": 0.9551,
|
| 329 |
+
"step": 19000
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 1.785591648925211,
|
| 333 |
+
"grad_norm": 3.698532819747925,
|
| 334 |
+
"learning_rate": 2.6697338403041827e-05,
|
| 335 |
+
"loss": 0.9585,
|
| 336 |
+
"step": 19500
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 1.8313760501797036,
|
| 340 |
+
"grad_norm": 1.4992170333862305,
|
| 341 |
+
"learning_rate": 2.6317110266159697e-05,
|
| 342 |
+
"loss": 0.951,
|
| 343 |
+
"step": 20000
|
| 344 |
+
},
|
| 345 |
+
{
|
| 346 |
+
"epoch": 1.8771604514341964,
|
| 347 |
+
"grad_norm": 3.0142011642456055,
|
| 348 |
+
"learning_rate": 2.5936882129277566e-05,
|
| 349 |
+
"loss": 0.9677,
|
| 350 |
+
"step": 20500
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 1.922944852688689,
|
| 354 |
+
"grad_norm": 1.9342625141143799,
|
| 355 |
+
"learning_rate": 2.5556653992395443e-05,
|
| 356 |
+
"loss": 0.9656,
|
| 357 |
+
"step": 21000
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 1.922944852688689,
|
| 361 |
+
"eval_loss": 0.3123963475227356,
|
| 362 |
+
"eval_runtime": 653.0667,
|
| 363 |
+
"eval_samples_per_second": 10.033,
|
| 364 |
+
"eval_steps_per_second": 10.033,
|
| 365 |
+
"eval_wer": 0.15938233272498875,
|
| 366 |
+
"step": 21000
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 1.9687292539431815,
|
| 370 |
+
"grad_norm": 5.628562927246094,
|
| 371 |
+
"learning_rate": 2.5177186311787077e-05,
|
| 372 |
+
"loss": 0.9511,
|
| 373 |
+
"step": 21500
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 2.0145136551976743,
|
| 377 |
+
"grad_norm": 2.930298089981079,
|
| 378 |
+
"learning_rate": 2.4796958174904946e-05,
|
| 379 |
+
"loss": 0.918,
|
| 380 |
+
"step": 22000
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 2.060298056452167,
|
| 384 |
+
"grad_norm": 1.2405227422714233,
|
| 385 |
+
"learning_rate": 2.4416730038022816e-05,
|
| 386 |
+
"loss": 0.9372,
|
| 387 |
+
"step": 22500
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 2.1060824577066595,
|
| 391 |
+
"grad_norm": 2.1558220386505127,
|
| 392 |
+
"learning_rate": 2.4036501901140686e-05,
|
| 393 |
+
"loss": 0.9673,
|
| 394 |
+
"step": 23000
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 2.151866858961152,
|
| 398 |
+
"grad_norm": 1.4096624851226807,
|
| 399 |
+
"learning_rate": 2.365703422053232e-05,
|
| 400 |
+
"loss": 0.9826,
|
| 401 |
+
"step": 23500
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 2.1976512602156446,
|
| 405 |
+
"grad_norm": 0.9921212196350098,
|
| 406 |
+
"learning_rate": 2.3276806083650192e-05,
|
| 407 |
+
"loss": 0.9261,
|
| 408 |
+
"step": 24000
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 2.1976512602156446,
|
| 412 |
+
"eval_loss": 0.30950331687927246,
|
| 413 |
+
"eval_runtime": 651.0611,
|
| 414 |
+
"eval_samples_per_second": 10.064,
|
| 415 |
+
"eval_steps_per_second": 10.064,
|
| 416 |
+
"eval_wer": 0.15439904244981156,
|
| 417 |
+
"step": 24000
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 2.243435661470137,
|
| 421 |
+
"grad_norm": 3.321803331375122,
|
| 422 |
+
"learning_rate": 2.2896577946768065e-05,
|
| 423 |
+
"loss": 0.8988,
|
| 424 |
+
"step": 24500
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 2.2892200627246297,
|
| 428 |
+
"grad_norm": 1.4012964963912964,
|
| 429 |
+
"learning_rate": 2.2516349809885935e-05,
|
| 430 |
+
"loss": 0.8717,
|
| 431 |
+
"step": 25000
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 2.3350044639791223,
|
| 435 |
+
"grad_norm": 1.9718306064605713,
|
| 436 |
+
"learning_rate": 2.2136121673003805e-05,
|
| 437 |
+
"loss": 0.9545,
|
| 438 |
+
"step": 25500
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 2.380788865233615,
|
| 442 |
+
"grad_norm": 2.5009524822235107,
|
| 443 |
+
"learning_rate": 2.1755893536121674e-05,
|
| 444 |
+
"loss": 0.9311,
|
| 445 |
+
"step": 26000
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 2.4265732664881074,
|
| 449 |
+
"grad_norm": 1.2410893440246582,
|
| 450 |
+
"learning_rate": 2.137642585551331e-05,
|
| 451 |
+
"loss": 0.8701,
|
| 452 |
+
"step": 26500
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 2.4723576677426,
|
| 456 |
+
"grad_norm": 1.5738617181777954,
|
| 457 |
+
"learning_rate": 2.099619771863118e-05,
|
| 458 |
+
"loss": 0.8938,
|
| 459 |
+
"step": 27000
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 2.4723576677426,
|
| 463 |
+
"eval_loss": 0.26886311173439026,
|
| 464 |
+
"eval_runtime": 654.1765,
|
| 465 |
+
"eval_samples_per_second": 10.016,
|
| 466 |
+
"eval_steps_per_second": 10.016,
|
| 467 |
+
"eval_wer": 0.15074304946552583,
|
| 468 |
+
"step": 27000
|
| 469 |
+
},
|
| 470 |
+
{
|
| 471 |
+
"epoch": 2.5181420689970926,
|
| 472 |
+
"grad_norm": 2.2281384468078613,
|
| 473 |
+
"learning_rate": 2.061596958174905e-05,
|
| 474 |
+
"loss": 0.8609,
|
| 475 |
+
"step": 27500
|
| 476 |
+
},
|
| 477 |
+
{
|
| 478 |
+
"epoch": 2.563926470251585,
|
| 479 |
+
"grad_norm": 2.347487449645996,
|
| 480 |
+
"learning_rate": 2.023574144486692e-05,
|
| 481 |
+
"loss": 0.9237,
|
| 482 |
+
"step": 28000
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 2.6097108715060777,
|
| 486 |
+
"grad_norm": 1.710715413093567,
|
| 487 |
+
"learning_rate": 1.9855513307984794e-05,
|
| 488 |
+
"loss": 0.8778,
|
| 489 |
+
"step": 28500
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 2.6554952727605707,
|
| 493 |
+
"grad_norm": 9.36032772064209,
|
| 494 |
+
"learning_rate": 1.947680608365019e-05,
|
| 495 |
+
"loss": 0.9092,
|
| 496 |
+
"step": 29000
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 2.701279674015063,
|
| 500 |
+
"grad_norm": 1.8656507730484009,
|
| 501 |
+
"learning_rate": 1.909657794676806e-05,
|
| 502 |
+
"loss": 0.8829,
|
| 503 |
+
"step": 29500
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 2.747064075269556,
|
| 507 |
+
"grad_norm": 0.8339200019836426,
|
| 508 |
+
"learning_rate": 1.8716349809885934e-05,
|
| 509 |
+
"loss": 0.8979,
|
| 510 |
+
"step": 30000
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 2.747064075269556,
|
| 514 |
+
"eval_loss": 0.2853592336177826,
|
| 515 |
+
"eval_runtime": 652.9254,
|
| 516 |
+
"eval_samples_per_second": 10.035,
|
| 517 |
+
"eval_steps_per_second": 10.035,
|
| 518 |
+
"eval_wer": 0.14718778886492379,
|
| 519 |
+
"step": 30000
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 2.792848476524048,
|
| 523 |
+
"grad_norm": 1.8189595937728882,
|
| 524 |
+
"learning_rate": 1.8336121673003804e-05,
|
| 525 |
+
"loss": 0.9216,
|
| 526 |
+
"step": 30500
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 2.838632877778541,
|
| 530 |
+
"grad_norm": 1.2487553358078003,
|
| 531 |
+
"learning_rate": 1.7955893536121673e-05,
|
| 532 |
+
"loss": 0.8714,
|
| 533 |
+
"step": 31000
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 2.8844172790330336,
|
| 537 |
+
"grad_norm": 2.8581480979919434,
|
| 538 |
+
"learning_rate": 1.757642585551331e-05,
|
| 539 |
+
"loss": 0.8675,
|
| 540 |
+
"step": 31500
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 2.930201680287526,
|
| 544 |
+
"grad_norm": 1.5087124109268188,
|
| 545 |
+
"learning_rate": 1.719619771863118e-05,
|
| 546 |
+
"loss": 0.9003,
|
| 547 |
+
"step": 32000
|
| 548 |
+
},
|
| 549 |
+
{
|
| 550 |
+
"epoch": 2.9759860815420187,
|
| 551 |
+
"grad_norm": 2.1860768795013428,
|
| 552 |
+
"learning_rate": 1.6816730038022814e-05,
|
| 553 |
+
"loss": 0.906,
|
| 554 |
+
"step": 32500
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 3.0217704827965113,
|
| 558 |
+
"grad_norm": 2.0595364570617676,
|
| 559 |
+
"learning_rate": 1.6436501901140687e-05,
|
| 560 |
+
"loss": 0.8624,
|
| 561 |
+
"step": 33000
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 3.0217704827965113,
|
| 565 |
+
"eval_loss": 0.282227486371994,
|
| 566 |
+
"eval_runtime": 652.2893,
|
| 567 |
+
"eval_samples_per_second": 10.045,
|
| 568 |
+
"eval_steps_per_second": 10.045,
|
| 569 |
+
"eval_wer": 0.14017800004740347,
|
| 570 |
+
"step": 33000
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 3.067554884051004,
|
| 574 |
+
"grad_norm": 2.31689453125,
|
| 575 |
+
"learning_rate": 1.6056273764258557e-05,
|
| 576 |
+
"loss": 0.8498,
|
| 577 |
+
"step": 33500
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 3.1133392853054964,
|
| 581 |
+
"grad_norm": 1.297116756439209,
|
| 582 |
+
"learning_rate": 1.5676045627376426e-05,
|
| 583 |
+
"loss": 0.8301,
|
| 584 |
+
"step": 34000
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 3.159123686559989,
|
| 588 |
+
"grad_norm": 1.9447718858718872,
|
| 589 |
+
"learning_rate": 1.52958174904943e-05,
|
| 590 |
+
"loss": 0.8445,
|
| 591 |
+
"step": 34500
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 3.2049080878144816,
|
| 595 |
+
"grad_norm": 1.8142869472503662,
|
| 596 |
+
"learning_rate": 1.4915589353612167e-05,
|
| 597 |
+
"loss": 0.8988,
|
| 598 |
+
"step": 35000
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 3.250692489068974,
|
| 602 |
+
"grad_norm": 26.81559944152832,
|
| 603 |
+
"learning_rate": 1.4535361216730039e-05,
|
| 604 |
+
"loss": 0.8358,
|
| 605 |
+
"step": 35500
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 3.2964768903234667,
|
| 609 |
+
"grad_norm": 1.4768437147140503,
|
| 610 |
+
"learning_rate": 1.4155133079847908e-05,
|
| 611 |
+
"loss": 0.8269,
|
| 612 |
+
"step": 36000
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 3.2964768903234667,
|
| 616 |
+
"eval_loss": 0.26364651322364807,
|
| 617 |
+
"eval_runtime": 653.8923,
|
| 618 |
+
"eval_samples_per_second": 10.02,
|
| 619 |
+
"eval_steps_per_second": 10.02,
|
| 620 |
+
"eval_wer": 0.13765376502097604,
|
| 621 |
+
"step": 36000
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"epoch": 3.3422612915779593,
|
| 625 |
+
"grad_norm": 3.271960496902466,
|
| 626 |
+
"learning_rate": 1.3774904942965781e-05,
|
| 627 |
+
"loss": 0.8389,
|
| 628 |
+
"step": 36500
|
| 629 |
+
},
|
| 630 |
+
{
|
| 631 |
+
"epoch": 3.388045692832452,
|
| 632 |
+
"grad_norm": 1.8337602615356445,
|
| 633 |
+
"learning_rate": 1.3395437262357415e-05,
|
| 634 |
+
"loss": 0.8295,
|
| 635 |
+
"step": 37000
|
| 636 |
+
},
|
| 637 |
+
{
|
| 638 |
+
"epoch": 3.4338300940869444,
|
| 639 |
+
"grad_norm": 2.3357372283935547,
|
| 640 |
+
"learning_rate": 1.301596958174905e-05,
|
| 641 |
+
"loss": 0.8646,
|
| 642 |
+
"step": 37500
|
| 643 |
+
},
|
| 644 |
+
{
|
| 645 |
+
"epoch": 3.479614495341437,
|
| 646 |
+
"grad_norm": 3.577268600463867,
|
| 647 |
+
"learning_rate": 1.2636501901140685e-05,
|
| 648 |
+
"loss": 0.844,
|
| 649 |
+
"step": 38000
|
| 650 |
+
},
|
| 651 |
+
{
|
| 652 |
+
"epoch": 3.52539889659593,
|
| 653 |
+
"grad_norm": 3.0567266941070557,
|
| 654 |
+
"learning_rate": 1.2256273764258558e-05,
|
| 655 |
+
"loss": 0.8212,
|
| 656 |
+
"step": 38500
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 3.571183297850422,
|
| 660 |
+
"grad_norm": 5.759117126464844,
|
| 661 |
+
"learning_rate": 1.1876045627376427e-05,
|
| 662 |
+
"loss": 0.85,
|
| 663 |
+
"step": 39000
|
| 664 |
+
},
|
| 665 |
+
{
|
| 666 |
+
"epoch": 3.571183297850422,
|
| 667 |
+
"eval_loss": 0.24618284404277802,
|
| 668 |
+
"eval_runtime": 648.8718,
|
| 669 |
+
"eval_samples_per_second": 10.098,
|
| 670 |
+
"eval_steps_per_second": 10.098,
|
| 671 |
+
"eval_wer": 0.13248678628143443,
|
| 672 |
+
"step": 39000
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 3.616967699104915,
|
| 676 |
+
"grad_norm": 1.7661995887756348,
|
| 677 |
+
"learning_rate": 1.1495817490494297e-05,
|
| 678 |
+
"loss": 0.8213,
|
| 679 |
+
"step": 39500
|
| 680 |
+
},
|
| 681 |
+
{
|
| 682 |
+
"epoch": 3.6627521003594077,
|
| 683 |
+
"grad_norm": 2.657759428024292,
|
| 684 |
+
"learning_rate": 1.1115589353612168e-05,
|
| 685 |
+
"loss": 0.8379,
|
| 686 |
+
"step": 40000
|
| 687 |
+
},
|
| 688 |
+
{
|
| 689 |
+
"epoch": 3.7085365016139002,
|
| 690 |
+
"grad_norm": 2.6153953075408936,
|
| 691 |
+
"learning_rate": 1.0735361216730038e-05,
|
| 692 |
+
"loss": 0.807,
|
| 693 |
+
"step": 40500
|
| 694 |
+
},
|
| 695 |
+
{
|
| 696 |
+
"epoch": 3.754320902868393,
|
| 697 |
+
"grad_norm": 4.843667507171631,
|
| 698 |
+
"learning_rate": 1.035513307984791e-05,
|
| 699 |
+
"loss": 0.8265,
|
| 700 |
+
"step": 41000
|
| 701 |
+
},
|
| 702 |
+
{
|
| 703 |
+
"epoch": 3.8001053041228854,
|
| 704 |
+
"grad_norm": 3.4964911937713623,
|
| 705 |
+
"learning_rate": 9.97490494296578e-06,
|
| 706 |
+
"loss": 0.85,
|
| 707 |
+
"step": 41500
|
| 708 |
+
},
|
| 709 |
+
{
|
| 710 |
+
"epoch": 3.845889705377378,
|
| 711 |
+
"grad_norm": 2.8312175273895264,
|
| 712 |
+
"learning_rate": 9.594676806083652e-06,
|
| 713 |
+
"loss": 0.8557,
|
| 714 |
+
"step": 42000
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 3.845889705377378,
|
| 718 |
+
"eval_loss": 0.24823951721191406,
|
| 719 |
+
"eval_runtime": 641.5128,
|
| 720 |
+
"eval_samples_per_second": 10.213,
|
| 721 |
+
"eval_steps_per_second": 10.213,
|
| 722 |
+
"eval_wer": 0.12771088620795903,
|
| 723 |
+
"step": 42000
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 3.8916741066318705,
|
| 727 |
+
"grad_norm": 2.185664653778076,
|
| 728 |
+
"learning_rate": 9.214448669201521e-06,
|
| 729 |
+
"loss": 0.8157,
|
| 730 |
+
"step": 42500
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 3.937458507886363,
|
| 734 |
+
"grad_norm": 3.490175724029541,
|
| 735 |
+
"learning_rate": 8.834220532319391e-06,
|
| 736 |
+
"loss": 0.8202,
|
| 737 |
+
"step": 43000
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 3.9832429091408557,
|
| 741 |
+
"grad_norm": 1.9320560693740845,
|
| 742 |
+
"learning_rate": 8.453992395437262e-06,
|
| 743 |
+
"loss": 0.8344,
|
| 744 |
+
"step": 43500
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 4.029027310395349,
|
| 748 |
+
"grad_norm": 1.6495819091796875,
|
| 749 |
+
"learning_rate": 8.073764258555134e-06,
|
| 750 |
+
"loss": 0.8167,
|
| 751 |
+
"step": 44000
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 4.074811711649841,
|
| 755 |
+
"grad_norm": 5.037049770355225,
|
| 756 |
+
"learning_rate": 7.694296577946768e-06,
|
| 757 |
+
"loss": 0.8033,
|
| 758 |
+
"step": 44500
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 4.120596112904334,
|
| 762 |
+
"grad_norm": 2.1924211978912354,
|
| 763 |
+
"learning_rate": 7.314828897338404e-06,
|
| 764 |
+
"loss": 0.8177,
|
| 765 |
+
"step": 45000
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 4.120596112904334,
|
| 769 |
+
"eval_loss": 0.2399299442768097,
|
| 770 |
+
"eval_runtime": 647.8262,
|
| 771 |
+
"eval_samples_per_second": 10.114,
|
| 772 |
+
"eval_steps_per_second": 10.114,
|
| 773 |
+
"eval_wer": 0.12562513332227251,
|
| 774 |
+
"step": 45000
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 4.166380514158826,
|
| 778 |
+
"grad_norm": 0.6167405843734741,
|
| 779 |
+
"learning_rate": 6.9346007604562745e-06,
|
| 780 |
+
"loss": 0.8205,
|
| 781 |
+
"step": 45500
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"epoch": 4.212164915413319,
|
| 785 |
+
"grad_norm": 1.5424410104751587,
|
| 786 |
+
"learning_rate": 6.554372623574144e-06,
|
| 787 |
+
"loss": 0.8096,
|
| 788 |
+
"step": 46000
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 4.257949316667811,
|
| 792 |
+
"grad_norm": 3.1014537811279297,
|
| 793 |
+
"learning_rate": 6.1741444866920154e-06,
|
| 794 |
+
"loss": 0.7904,
|
| 795 |
+
"step": 46500
|
| 796 |
+
},
|
| 797 |
+
{
|
| 798 |
+
"epoch": 4.303733717922304,
|
| 799 |
+
"grad_norm": 1.1060818433761597,
|
| 800 |
+
"learning_rate": 5.794676806083651e-06,
|
| 801 |
+
"loss": 0.7818,
|
| 802 |
+
"step": 47000
|
| 803 |
+
},
|
| 804 |
+
{
|
| 805 |
+
"epoch": 4.349518119176796,
|
| 806 |
+
"grad_norm": 1.345553994178772,
|
| 807 |
+
"learning_rate": 5.414448669201522e-06,
|
| 808 |
+
"loss": 0.8144,
|
| 809 |
+
"step": 47500
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 4.395302520431289,
|
| 813 |
+
"grad_norm": 2.172055721282959,
|
| 814 |
+
"learning_rate": 5.034220532319392e-06,
|
| 815 |
+
"loss": 0.7933,
|
| 816 |
+
"step": 48000
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 4.395302520431289,
|
| 820 |
+
"eval_loss": 0.2388318032026291,
|
| 821 |
+
"eval_runtime": 640.8224,
|
| 822 |
+
"eval_samples_per_second": 10.224,
|
| 823 |
+
"eval_steps_per_second": 10.224,
|
| 824 |
+
"eval_wer": 0.12215282880235122,
|
| 825 |
+
"step": 48000
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 4.441086921685781,
|
| 829 |
+
"grad_norm": 3.54559063911438,
|
| 830 |
+
"learning_rate": 4.653992395437263e-06,
|
| 831 |
+
"loss": 0.814,
|
| 832 |
+
"step": 48500
|
| 833 |
+
},
|
| 834 |
+
{
|
| 835 |
+
"epoch": 4.486871322940274,
|
| 836 |
+
"grad_norm": 3.8439674377441406,
|
| 837 |
+
"learning_rate": 4.273764258555134e-06,
|
| 838 |
+
"loss": 0.8641,
|
| 839 |
+
"step": 49000
|
| 840 |
+
},
|
| 841 |
+
{
|
| 842 |
+
"epoch": 4.5326557241947665,
|
| 843 |
+
"grad_norm": 1.967161774635315,
|
| 844 |
+
"learning_rate": 3.893536121673004e-06,
|
| 845 |
+
"loss": 0.7962,
|
| 846 |
+
"step": 49500
|
| 847 |
+
},
|
| 848 |
+
{
|
| 849 |
+
"epoch": 4.5784401254492595,
|
| 850 |
+
"grad_norm": 1.7992150783538818,
|
| 851 |
+
"learning_rate": 3.5133079847908747e-06,
|
| 852 |
+
"loss": 0.7871,
|
| 853 |
+
"step": 50000
|
| 854 |
+
},
|
| 855 |
+
{
|
| 856 |
+
"epoch": 4.6242245267037525,
|
| 857 |
+
"grad_norm": 1.4352868795394897,
|
| 858 |
+
"learning_rate": 3.1330798479087456e-06,
|
| 859 |
+
"loss": 0.839,
|
| 860 |
+
"step": 50500
|
| 861 |
+
},
|
| 862 |
+
{
|
| 863 |
+
"epoch": 4.670008927958245,
|
| 864 |
+
"grad_norm": 1.4436633586883545,
|
| 865 |
+
"learning_rate": 2.7536121673003807e-06,
|
| 866 |
+
"loss": 0.7864,
|
| 867 |
+
"step": 51000
|
| 868 |
+
},
|
| 869 |
+
{
|
| 870 |
+
"epoch": 4.670008927958245,
|
| 871 |
+
"eval_loss": 0.237684428691864,
|
| 872 |
+
"eval_runtime": 642.8597,
|
| 873 |
+
"eval_samples_per_second": 10.192,
|
| 874 |
+
"eval_steps_per_second": 10.192,
|
| 875 |
+
"eval_wer": 0.12045815458273092,
|
| 876 |
+
"step": 51000
|
| 877 |
+
},
|
| 878 |
+
{
|
| 879 |
+
"epoch": 4.715793329212737,
|
| 880 |
+
"grad_norm": 6.0789055824279785,
|
| 881 |
+
"learning_rate": 2.373384030418251e-06,
|
| 882 |
+
"loss": 0.7787,
|
| 883 |
+
"step": 51500
|
| 884 |
+
},
|
| 885 |
+
{
|
| 886 |
+
"epoch": 4.76157773046723,
|
| 887 |
+
"grad_norm": 3.2274794578552246,
|
| 888 |
+
"learning_rate": 1.993916349809886e-06,
|
| 889 |
+
"loss": 0.8234,
|
| 890 |
+
"step": 52000
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 4.807362131721723,
|
| 894 |
+
"grad_norm": 2.8651950359344482,
|
| 895 |
+
"learning_rate": 1.6136882129277568e-06,
|
| 896 |
+
"loss": 0.8,
|
| 897 |
+
"step": 52500
|
| 898 |
+
},
|
| 899 |
+
{
|
| 900 |
+
"epoch": 4.853146532976215,
|
| 901 |
+
"grad_norm": 1.4273629188537598,
|
| 902 |
+
"learning_rate": 1.2334600760456275e-06,
|
| 903 |
+
"loss": 0.7861,
|
| 904 |
+
"step": 53000
|
| 905 |
+
},
|
| 906 |
+
{
|
| 907 |
+
"epoch": 4.898930934230708,
|
| 908 |
+
"grad_norm": 1.4547092914581299,
|
| 909 |
+
"learning_rate": 8.532319391634982e-07,
|
| 910 |
+
"loss": 0.7801,
|
| 911 |
+
"step": 53500
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"epoch": 4.9447153354852,
|
| 915 |
+
"grad_norm": 2.2551023960113525,
|
| 916 |
+
"learning_rate": 4.737642585551331e-07,
|
| 917 |
+
"loss": 0.8371,
|
| 918 |
+
"step": 54000
|
| 919 |
+
},
|
| 920 |
+
{
|
| 921 |
+
"epoch": 4.9447153354852,
|
| 922 |
+
"eval_loss": 0.23816268146038055,
|
| 923 |
+
"eval_runtime": 641.5849,
|
| 924 |
+
"eval_samples_per_second": 10.212,
|
| 925 |
+
"eval_steps_per_second": 10.212,
|
| 926 |
+
"eval_wer": 0.11996634353298097,
|
| 927 |
+
"step": 54000
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 4.990499736739693,
|
| 931 |
+
"grad_norm": 1.8520026206970215,
|
| 932 |
+
"learning_rate": 9.353612167300382e-08,
|
| 933 |
+
"loss": 0.7373,
|
| 934 |
+
"step": 54500
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 5.036627521003594,
|
| 938 |
+
"grad_norm": 5.179400444030762,
|
| 939 |
+
"learning_rate": 2.0232462686567166e-05,
|
| 940 |
+
"loss": 0.836,
|
| 941 |
+
"step": 55000
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 5.082411922258086,
|
| 945 |
+
"grad_norm": 7.242797374725342,
|
| 946 |
+
"learning_rate": 2.004589552238806e-05,
|
| 947 |
+
"loss": 0.8493,
|
| 948 |
+
"step": 55500
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 5.128196323512579,
|
| 952 |
+
"grad_norm": 3.6222658157348633,
|
| 953 |
+
"learning_rate": 1.9859328358208957e-05,
|
| 954 |
+
"loss": 0.8105,
|
| 955 |
+
"step": 56000
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 5.1739807247670715,
|
| 959 |
+
"grad_norm": 2.839559316635132,
|
| 960 |
+
"learning_rate": 1.9672761194029853e-05,
|
| 961 |
+
"loss": 0.8075,
|
| 962 |
+
"step": 56500
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 5.2197651260215645,
|
| 966 |
+
"grad_norm": 2.1728529930114746,
|
| 967 |
+
"learning_rate": 1.9486940298507463e-05,
|
| 968 |
+
"loss": 0.7846,
|
| 969 |
+
"step": 57000
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 5.2197651260215645,
|
| 973 |
+
"eval_loss": 0.24261941015720367,
|
| 974 |
+
"eval_runtime": 650.6489,
|
| 975 |
+
"eval_samples_per_second": 10.07,
|
| 976 |
+
"eval_steps_per_second": 10.07,
|
| 977 |
+
"eval_wer": 0.13190609371666942,
|
| 978 |
+
"step": 57000
|
| 979 |
+
},
|
| 980 |
+
{
|
| 981 |
+
"epoch": 5.2655495272760575,
|
| 982 |
+
"grad_norm": 1.945101022720337,
|
| 983 |
+
"learning_rate": 1.9300373134328362e-05,
|
| 984 |
+
"loss": 0.8194,
|
| 985 |
+
"step": 57500
|
| 986 |
+
},
|
| 987 |
+
{
|
| 988 |
+
"epoch": 5.31133392853055,
|
| 989 |
+
"grad_norm": 1.497315764427185,
|
| 990 |
+
"learning_rate": 1.9113805970149254e-05,
|
| 991 |
+
"loss": 0.8367,
|
| 992 |
+
"step": 58000
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 5.357118329785042,
|
| 996 |
+
"grad_norm": 3.0750606060028076,
|
| 997 |
+
"learning_rate": 1.8927238805970153e-05,
|
| 998 |
+
"loss": 0.8442,
|
| 999 |
+
"step": 58500
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 5.402902731039535,
|
| 1003 |
+
"grad_norm": 2.4695701599121094,
|
| 1004 |
+
"learning_rate": 1.8740671641791046e-05,
|
| 1005 |
+
"loss": 0.8124,
|
| 1006 |
+
"step": 59000
|
| 1007 |
+
},
|
| 1008 |
+
{
|
| 1009 |
+
"epoch": 5.448687132294028,
|
| 1010 |
+
"grad_norm": 1.8706340789794922,
|
| 1011 |
+
"learning_rate": 1.85544776119403e-05,
|
| 1012 |
+
"loss": 0.775,
|
| 1013 |
+
"step": 59500
|
| 1014 |
+
},
|
| 1015 |
+
{
|
| 1016 |
+
"epoch": 5.49447153354852,
|
| 1017 |
+
"grad_norm": 5.028220176696777,
|
| 1018 |
+
"learning_rate": 1.8367910447761194e-05,
|
| 1019 |
+
"loss": 0.808,
|
| 1020 |
+
"step": 60000
|
| 1021 |
+
},
|
| 1022 |
+
{
|
| 1023 |
+
"epoch": 5.49447153354852,
|
| 1024 |
+
"eval_loss": 0.24327826499938965,
|
| 1025 |
+
"eval_runtime": 650.6547,
|
| 1026 |
+
"eval_samples_per_second": 10.07,
|
| 1027 |
+
"eval_steps_per_second": 10.07,
|
| 1028 |
+
"eval_wer": 0.13021734493138348,
|
| 1029 |
+
"step": 60000
|
| 1030 |
+
},
|
| 1031 |
+
{
|
| 1032 |
+
"epoch": 5.540255934803013,
|
| 1033 |
+
"grad_norm": 1.9498703479766846,
|
| 1034 |
+
"learning_rate": 1.818134328358209e-05,
|
| 1035 |
+
"loss": 0.7813,
|
| 1036 |
+
"step": 60500
|
| 1037 |
+
},
|
| 1038 |
+
{
|
| 1039 |
+
"epoch": 5.586040336057505,
|
| 1040 |
+
"grad_norm": 3.7338151931762695,
|
| 1041 |
+
"learning_rate": 1.7994776119402986e-05,
|
| 1042 |
+
"loss": 0.7698,
|
| 1043 |
+
"step": 61000
|
| 1044 |
+
},
|
| 1045 |
+
{
|
| 1046 |
+
"epoch": 5.631824737311998,
|
| 1047 |
+
"grad_norm": 3.364290475845337,
|
| 1048 |
+
"learning_rate": 1.780820895522388e-05,
|
| 1049 |
+
"loss": 0.7783,
|
| 1050 |
+
"step": 61500
|
| 1051 |
+
},
|
| 1052 |
+
{
|
| 1053 |
+
"epoch": 5.67760913856649,
|
| 1054 |
+
"grad_norm": 3.4009501934051514,
|
| 1055 |
+
"learning_rate": 1.7622014925373137e-05,
|
| 1056 |
+
"loss": 0.7652,
|
| 1057 |
+
"step": 62000
|
| 1058 |
+
},
|
| 1059 |
+
{
|
| 1060 |
+
"epoch": 5.723393539820983,
|
| 1061 |
+
"grad_norm": 2.9030606746673584,
|
| 1062 |
+
"learning_rate": 1.743544776119403e-05,
|
| 1063 |
+
"loss": 0.7946,
|
| 1064 |
+
"step": 62500
|
| 1065 |
+
},
|
| 1066 |
+
{
|
| 1067 |
+
"epoch": 5.769177941075475,
|
| 1068 |
+
"grad_norm": 1.3028395175933838,
|
| 1069 |
+
"learning_rate": 1.724888059701493e-05,
|
| 1070 |
+
"loss": 0.7456,
|
| 1071 |
+
"step": 63000
|
| 1072 |
+
},
|
| 1073 |
+
{
|
| 1074 |
+
"epoch": 5.769177941075475,
|
| 1075 |
+
"eval_loss": 0.2477836161851883,
|
| 1076 |
+
"eval_runtime": 648.6941,
|
| 1077 |
+
"eval_samples_per_second": 10.1,
|
| 1078 |
+
"eval_steps_per_second": 10.1,
|
| 1079 |
+
"eval_wer": 0.1254769974639141,
|
| 1080 |
+
"step": 63000
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 5.814962342329968,
|
| 1084 |
+
"grad_norm": 1.8734581470489502,
|
| 1085 |
+
"learning_rate": 1.706231343283582e-05,
|
| 1086 |
+
"loss": 0.7893,
|
| 1087 |
+
"step": 63500
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 5.8607467435844605,
|
| 1091 |
+
"grad_norm": 2.084261178970337,
|
| 1092 |
+
"learning_rate": 1.6876119402985077e-05,
|
| 1093 |
+
"loss": 0.7331,
|
| 1094 |
+
"step": 64000
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 5.9065311448389535,
|
| 1098 |
+
"grad_norm": 2.031810760498047,
|
| 1099 |
+
"learning_rate": 1.668955223880597e-05,
|
| 1100 |
+
"loss": 0.747,
|
| 1101 |
+
"step": 64500
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 5.952315546093446,
|
| 1105 |
+
"grad_norm": 4.709017276763916,
|
| 1106 |
+
"learning_rate": 1.650298507462687e-05,
|
| 1107 |
+
"loss": 0.7473,
|
| 1108 |
+
"step": 65000
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 5.998099947347939,
|
| 1112 |
+
"grad_norm": 3.2480199337005615,
|
| 1113 |
+
"learning_rate": 1.631641791044776e-05,
|
| 1114 |
+
"loss": 0.7195,
|
| 1115 |
+
"step": 65500
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 6.043884348602431,
|
| 1119 |
+
"grad_norm": 2.1403627395629883,
|
| 1120 |
+
"learning_rate": 1.6129850746268657e-05,
|
| 1121 |
+
"loss": 0.7217,
|
| 1122 |
+
"step": 66000
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 6.043884348602431,
|
| 1126 |
+
"eval_loss": 0.24559348821640015,
|
| 1127 |
+
"eval_runtime": 650.412,
|
| 1128 |
+
"eval_samples_per_second": 10.074,
|
| 1129 |
+
"eval_steps_per_second": 10.074,
|
| 1130 |
+
"eval_wer": 0.12621175132137186,
|
| 1131 |
+
"step": 66000
|
| 1132 |
+
},
|
| 1133 |
+
{
|
| 1134 |
+
"epoch": 6.089668749856924,
|
| 1135 |
+
"grad_norm": 2.90395188331604,
|
| 1136 |
+
"learning_rate": 1.5943283582089553e-05,
|
| 1137 |
+
"loss": 0.7285,
|
| 1138 |
+
"step": 66500
|
| 1139 |
+
},
|
| 1140 |
+
{
|
| 1141 |
+
"epoch": 6.135453151111417,
|
| 1142 |
+
"grad_norm": 2.9291510581970215,
|
| 1143 |
+
"learning_rate": 1.5757089552238805e-05,
|
| 1144 |
+
"loss": 0.7324,
|
| 1145 |
+
"step": 67000
|
| 1146 |
+
},
|
| 1147 |
+
{
|
| 1148 |
+
"epoch": 6.181237552365909,
|
| 1149 |
+
"grad_norm": 12.322357177734375,
|
| 1150 |
+
"learning_rate": 1.5570522388059705e-05,
|
| 1151 |
+
"loss": 0.7198,
|
| 1152 |
+
"step": 67500
|
| 1153 |
+
},
|
| 1154 |
+
{
|
| 1155 |
+
"epoch": 6.227021953620402,
|
| 1156 |
+
"grad_norm": 4.177962303161621,
|
| 1157 |
+
"learning_rate": 1.5383955223880597e-05,
|
| 1158 |
+
"loss": 0.7206,
|
| 1159 |
+
"step": 68000
|
| 1160 |
+
},
|
| 1161 |
+
{
|
| 1162 |
+
"epoch": 6.272806354874894,
|
| 1163 |
+
"grad_norm": 1.2690945863723755,
|
| 1164 |
+
"learning_rate": 1.5197388059701494e-05,
|
| 1165 |
+
"loss": 0.729,
|
| 1166 |
+
"step": 68500
|
| 1167 |
+
},
|
| 1168 |
+
{
|
| 1169 |
+
"epoch": 6.318590756129387,
|
| 1170 |
+
"grad_norm": 1.8948352336883545,
|
| 1171 |
+
"learning_rate": 1.5010820895522389e-05,
|
| 1172 |
+
"loss": 0.7115,
|
| 1173 |
+
"step": 69000
|
| 1174 |
+
},
|
| 1175 |
+
{
|
| 1176 |
+
"epoch": 6.318590756129387,
|
| 1177 |
+
"eval_loss": 0.2540062367916107,
|
| 1178 |
+
"eval_runtime": 648.0874,
|
| 1179 |
+
"eval_samples_per_second": 10.11,
|
| 1180 |
+
"eval_steps_per_second": 10.11,
|
| 1181 |
+
"eval_wer": 0.12306534568983907,
|
| 1182 |
+
"step": 69000
|
| 1183 |
+
},
|
| 1184 |
+
{
|
| 1185 |
+
"epoch": 6.364375157383879,
|
| 1186 |
+
"grad_norm": 2.9900403022766113,
|
| 1187 |
+
"learning_rate": 1.4824253731343286e-05,
|
| 1188 |
+
"loss": 0.7681,
|
| 1189 |
+
"step": 69500
|
| 1190 |
+
},
|
| 1191 |
+
{
|
| 1192 |
+
"epoch": 6.410159558638372,
|
| 1193 |
+
"grad_norm": 3.5702733993530273,
|
| 1194 |
+
"learning_rate": 1.4638059701492537e-05,
|
| 1195 |
+
"loss": 0.7142,
|
| 1196 |
+
"step": 70000
|
| 1197 |
+
},
|
| 1198 |
+
{
|
| 1199 |
+
"epoch": 6.455943959892864,
|
| 1200 |
+
"grad_norm": 2.3290789127349854,
|
| 1201 |
+
"learning_rate": 1.4451492537313434e-05,
|
| 1202 |
+
"loss": 0.71,
|
| 1203 |
+
"step": 70500
|
| 1204 |
+
},
|
| 1205 |
+
{
|
| 1206 |
+
"epoch": 6.501728361147357,
|
| 1207 |
+
"grad_norm": 6.307437896728516,
|
| 1208 |
+
"learning_rate": 1.4264925373134328e-05,
|
| 1209 |
+
"loss": 0.6944,
|
| 1210 |
+
"step": 71000
|
| 1211 |
+
},
|
| 1212 |
+
{
|
| 1213 |
+
"epoch": 6.547512762401849,
|
| 1214 |
+
"grad_norm": 5.269557952880859,
|
| 1215 |
+
"learning_rate": 1.4078358208955226e-05,
|
| 1216 |
+
"loss": 0.7638,
|
| 1217 |
+
"step": 71500
|
| 1218 |
+
},
|
| 1219 |
+
{
|
| 1220 |
+
"epoch": 6.593297163656342,
|
| 1221 |
+
"grad_norm": 1.6152143478393555,
|
| 1222 |
+
"learning_rate": 1.389179104477612e-05,
|
| 1223 |
+
"loss": 0.6701,
|
| 1224 |
+
"step": 72000
|
| 1225 |
+
},
|
| 1226 |
+
{
|
| 1227 |
+
"epoch": 6.593297163656342,
|
| 1228 |
+
"eval_loss": 0.2436157912015915,
|
| 1229 |
+
"eval_runtime": 649.7342,
|
| 1230 |
+
"eval_samples_per_second": 10.084,
|
| 1231 |
+
"eval_steps_per_second": 10.084,
|
| 1232 |
+
"eval_wer": 0.12396008627432391,
|
| 1233 |
+
"step": 72000
|
| 1234 |
+
},
|
| 1235 |
+
{
|
| 1236 |
+
"epoch": 6.639081564910835,
|
| 1237 |
+
"grad_norm": 2.4076077938079834,
|
| 1238 |
+
"learning_rate": 1.3705223880597016e-05,
|
| 1239 |
+
"loss": 0.7058,
|
| 1240 |
+
"step": 72500
|
| 1241 |
+
},
|
| 1242 |
+
{
|
| 1243 |
+
"epoch": 6.684865966165328,
|
| 1244 |
+
"grad_norm": 3.9522953033447266,
|
| 1245 |
+
"learning_rate": 1.351865671641791e-05,
|
| 1246 |
+
"loss": 0.6958,
|
| 1247 |
+
"step": 73000
|
| 1248 |
+
},
|
| 1249 |
+
{
|
| 1250 |
+
"epoch": 6.73065036741982,
|
| 1251 |
+
"grad_norm": 1.8242387771606445,
|
| 1252 |
+
"learning_rate": 1.3332089552238807e-05,
|
| 1253 |
+
"loss": 0.7166,
|
| 1254 |
+
"step": 73500
|
| 1255 |
+
},
|
| 1256 |
+
{
|
| 1257 |
+
"epoch": 6.776434768674313,
|
| 1258 |
+
"grad_norm": 4.560197353363037,
|
| 1259 |
+
"learning_rate": 1.3145522388059701e-05,
|
| 1260 |
+
"loss": 0.6946,
|
| 1261 |
+
"step": 74000
|
| 1262 |
+
},
|
| 1263 |
+
{
|
| 1264 |
+
"epoch": 6.822219169928806,
|
| 1265 |
+
"grad_norm": 2.9841842651367188,
|
| 1266 |
+
"learning_rate": 1.2958955223880599e-05,
|
| 1267 |
+
"loss": 0.7179,
|
| 1268 |
+
"step": 74500
|
| 1269 |
+
},
|
| 1270 |
+
{
|
| 1271 |
+
"epoch": 6.868003571183298,
|
| 1272 |
+
"grad_norm": 1.7246359586715698,
|
| 1273 |
+
"learning_rate": 1.2772761194029852e-05,
|
| 1274 |
+
"loss": 0.6966,
|
| 1275 |
+
"step": 75000
|
| 1276 |
+
},
|
| 1277 |
+
{
|
| 1278 |
+
"epoch": 6.868003571183298,
|
| 1279 |
+
"eval_loss": 0.23714399337768555,
|
| 1280 |
+
"eval_runtime": 653.4649,
|
| 1281 |
+
"eval_samples_per_second": 10.027,
|
| 1282 |
+
"eval_steps_per_second": 10.027,
|
| 1283 |
+
"eval_wer": 0.12096774193548387,
|
| 1284 |
+
"step": 75000
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 6.91378797243779,
|
| 1288 |
+
"grad_norm": 3.9530229568481445,
|
| 1289 |
+
"learning_rate": 1.2586194029850747e-05,
|
| 1290 |
+
"loss": 0.6542,
|
| 1291 |
+
"step": 75500
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 6.959572373692283,
|
| 1295 |
+
"grad_norm": 4.2303690910339355,
|
| 1296 |
+
"learning_rate": 1.2399626865671643e-05,
|
| 1297 |
+
"loss": 0.683,
|
| 1298 |
+
"step": 76000
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 7.005356774946776,
|
| 1302 |
+
"grad_norm": 1.5437530279159546,
|
| 1303 |
+
"learning_rate": 1.2213059701492537e-05,
|
| 1304 |
+
"loss": 0.6415,
|
| 1305 |
+
"step": 76500
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 7.051141176201268,
|
| 1309 |
+
"grad_norm": 2.246727466583252,
|
| 1310 |
+
"learning_rate": 1.2026492537313435e-05,
|
| 1311 |
+
"loss": 0.6831,
|
| 1312 |
+
"step": 77000
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 7.096925577455761,
|
| 1316 |
+
"grad_norm": 5.2752580642700195,
|
| 1317 |
+
"learning_rate": 1.1839925373134329e-05,
|
| 1318 |
+
"loss": 0.6666,
|
| 1319 |
+
"step": 77500
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 7.142709978710253,
|
| 1323 |
+
"grad_norm": 1.453157663345337,
|
| 1324 |
+
"learning_rate": 1.1653358208955226e-05,
|
| 1325 |
+
"loss": 0.7056,
|
| 1326 |
+
"step": 78000
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 7.142709978710253,
|
| 1330 |
+
"eval_loss": 0.23014602065086365,
|
| 1331 |
+
"eval_runtime": 648.8331,
|
| 1332 |
+
"eval_samples_per_second": 10.098,
|
| 1333 |
+
"eval_steps_per_second": 10.098,
|
| 1334 |
+
"eval_wer": 0.11875755492877628,
|
| 1335 |
+
"step": 78000
|
| 1336 |
+
},
|
| 1337 |
+
{
|
| 1338 |
+
"epoch": 7.188494379964746,
|
| 1339 |
+
"grad_norm": 1.6704216003417969,
|
| 1340 |
+
"learning_rate": 1.1467164179104479e-05,
|
| 1341 |
+
"loss": 0.7214,
|
| 1342 |
+
"step": 78500
|
| 1343 |
+
},
|
| 1344 |
+
{
|
| 1345 |
+
"epoch": 7.234278781219238,
|
| 1346 |
+
"grad_norm": 2.2454934120178223,
|
| 1347 |
+
"learning_rate": 1.1280597014925375e-05,
|
| 1348 |
+
"loss": 0.6921,
|
| 1349 |
+
"step": 79000
|
| 1350 |
+
},
|
| 1351 |
+
{
|
| 1352 |
+
"epoch": 7.280063182473731,
|
| 1353 |
+
"grad_norm": 0.8851762413978577,
|
| 1354 |
+
"learning_rate": 1.1094029850746269e-05,
|
| 1355 |
+
"loss": 0.6605,
|
| 1356 |
+
"step": 79500
|
| 1357 |
+
},
|
| 1358 |
+
{
|
| 1359 |
+
"epoch": 7.3258475837282235,
|
| 1360 |
+
"grad_norm": 2.482797861099243,
|
| 1361 |
+
"learning_rate": 1.0907462686567165e-05,
|
| 1362 |
+
"loss": 0.6383,
|
| 1363 |
+
"step": 80000
|
| 1364 |
+
},
|
| 1365 |
+
{
|
| 1366 |
+
"epoch": 7.3716319849827165,
|
| 1367 |
+
"grad_norm": 2.4076988697052,
|
| 1368 |
+
"learning_rate": 1.072089552238806e-05,
|
| 1369 |
+
"loss": 0.6638,
|
| 1370 |
+
"step": 80500
|
| 1371 |
+
},
|
| 1372 |
+
{
|
| 1373 |
+
"epoch": 7.417416386237209,
|
| 1374 |
+
"grad_norm": 2.0605874061584473,
|
| 1375 |
+
"learning_rate": 1.0534328358208956e-05,
|
| 1376 |
+
"loss": 0.6654,
|
| 1377 |
+
"step": 81000
|
| 1378 |
+
},
|
| 1379 |
+
{
|
| 1380 |
+
"epoch": 7.417416386237209,
|
| 1381 |
+
"eval_loss": 0.24034035205841064,
|
| 1382 |
+
"eval_runtime": 650.4954,
|
| 1383 |
+
"eval_samples_per_second": 10.072,
|
| 1384 |
+
"eval_steps_per_second": 10.072,
|
| 1385 |
+
"eval_wer": 0.11891754165580337,
|
| 1386 |
+
"step": 81000
|
| 1387 |
+
},
|
| 1388 |
+
{
|
| 1389 |
+
"epoch": 7.463200787491702,
|
| 1390 |
+
"grad_norm": 1.975950002670288,
|
| 1391 |
+
"learning_rate": 1.034776119402985e-05,
|
| 1392 |
+
"loss": 0.6744,
|
| 1393 |
+
"step": 81500
|
| 1394 |
+
},
|
| 1395 |
+
{
|
| 1396 |
+
"epoch": 7.508985188746194,
|
| 1397 |
+
"grad_norm": 1.52113676071167,
|
| 1398 |
+
"learning_rate": 1.0161940298507465e-05,
|
| 1399 |
+
"loss": 0.6832,
|
| 1400 |
+
"step": 82000
|
| 1401 |
+
},
|
| 1402 |
+
{
|
| 1403 |
+
"epoch": 7.554769590000687,
|
| 1404 |
+
"grad_norm": 2.152534246444702,
|
| 1405 |
+
"learning_rate": 9.975373134328359e-06,
|
| 1406 |
+
"loss": 0.6719,
|
| 1407 |
+
"step": 82500
|
| 1408 |
+
},
|
| 1409 |
+
{
|
| 1410 |
+
"epoch": 7.600553991255179,
|
| 1411 |
+
"grad_norm": 1.5796542167663574,
|
| 1412 |
+
"learning_rate": 9.788805970149254e-06,
|
| 1413 |
+
"loss": 0.6515,
|
| 1414 |
+
"step": 83000
|
| 1415 |
+
},
|
| 1416 |
+
{
|
| 1417 |
+
"epoch": 7.646338392509672,
|
| 1418 |
+
"grad_norm": 1.5929287672042847,
|
| 1419 |
+
"learning_rate": 9.60223880597015e-06,
|
| 1420 |
+
"loss": 0.666,
|
| 1421 |
+
"step": 83500
|
| 1422 |
+
},
|
| 1423 |
+
{
|
| 1424 |
+
"epoch": 7.692122793764165,
|
| 1425 |
+
"grad_norm": 1.7882949113845825,
|
| 1426 |
+
"learning_rate": 9.415671641791046e-06,
|
| 1427 |
+
"loss": 0.6339,
|
| 1428 |
+
"step": 84000
|
| 1429 |
+
},
|
| 1430 |
+
{
|
| 1431 |
+
"epoch": 7.692122793764165,
|
| 1432 |
+
"eval_loss": 0.23006904125213623,
|
| 1433 |
+
"eval_runtime": 649.7568,
|
| 1434 |
+
"eval_samples_per_second": 10.084,
|
| 1435 |
+
"eval_steps_per_second": 10.084,
|
| 1436 |
+
"eval_wer": 0.11488232087412008,
|
| 1437 |
+
"step": 84000
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 7.737907195018657,
|
| 1441 |
+
"grad_norm": 1.4368743896484375,
|
| 1442 |
+
"learning_rate": 9.22910447761194e-06,
|
| 1443 |
+
"loss": 0.6861,
|
| 1444 |
+
"step": 84500
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 7.78369159627315,
|
| 1448 |
+
"grad_norm": 2.0316922664642334,
|
| 1449 |
+
"learning_rate": 9.042537313432836e-06,
|
| 1450 |
+
"loss": 0.6513,
|
| 1451 |
+
"step": 85000
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 7.829475997527642,
|
| 1455 |
+
"grad_norm": 1.2255696058273315,
|
| 1456 |
+
"learning_rate": 8.855970149253732e-06,
|
| 1457 |
+
"loss": 0.6336,
|
| 1458 |
+
"step": 85500
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 7.875260398782135,
|
| 1462 |
+
"grad_norm": 2.4979262351989746,
|
| 1463 |
+
"learning_rate": 8.669776119402986e-06,
|
| 1464 |
+
"loss": 0.639,
|
| 1465 |
+
"step": 86000
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 7.921044800036627,
|
| 1469 |
+
"grad_norm": 2.6587982177734375,
|
| 1470 |
+
"learning_rate": 8.483208955223882e-06,
|
| 1471 |
+
"loss": 0.6506,
|
| 1472 |
+
"step": 86500
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 7.96682920129112,
|
| 1476 |
+
"grad_norm": 2.314595937728882,
|
| 1477 |
+
"learning_rate": 8.296641791044778e-06,
|
| 1478 |
+
"loss": 0.6649,
|
| 1479 |
+
"step": 87000
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 7.96682920129112,
|
| 1483 |
+
"eval_loss": 0.24233344197273254,
|
| 1484 |
+
"eval_runtime": 654.1461,
|
| 1485 |
+
"eval_samples_per_second": 10.016,
|
| 1486 |
+
"eval_steps_per_second": 10.016,
|
| 1487 |
+
"eval_wer": 0.11500675499514114,
|
| 1488 |
+
"step": 87000
|
| 1489 |
+
},
|
| 1490 |
+
{
|
| 1491 |
+
"epoch": 8.012613602545613,
|
| 1492 |
+
"grad_norm": 2.341566562652588,
|
| 1493 |
+
"learning_rate": 8.110074626865673e-06,
|
| 1494 |
+
"loss": 0.6722,
|
| 1495 |
+
"step": 87500
|
| 1496 |
+
},
|
| 1497 |
+
{
|
| 1498 |
+
"epoch": 8.058398003800106,
|
| 1499 |
+
"grad_norm": 3.6063222885131836,
|
| 1500 |
+
"learning_rate": 7.923507462686567e-06,
|
| 1501 |
+
"loss": 0.6567,
|
| 1502 |
+
"step": 88000
|
| 1503 |
+
},
|
| 1504 |
+
{
|
| 1505 |
+
"epoch": 8.104182405054598,
|
| 1506 |
+
"grad_norm": 3.134598970413208,
|
| 1507 |
+
"learning_rate": 7.736940298507463e-06,
|
| 1508 |
+
"loss": 0.6455,
|
| 1509 |
+
"step": 88500
|
| 1510 |
+
},
|
| 1511 |
+
{
|
| 1512 |
+
"epoch": 8.14996680630909,
|
| 1513 |
+
"grad_norm": 2.014883041381836,
|
| 1514 |
+
"learning_rate": 7.550373134328359e-06,
|
| 1515 |
+
"loss": 0.6618,
|
| 1516 |
+
"step": 89000
|
| 1517 |
+
},
|
| 1518 |
+
{
|
| 1519 |
+
"epoch": 8.195751207563584,
|
| 1520 |
+
"grad_norm": 2.0325255393981934,
|
| 1521 |
+
"learning_rate": 7.363805970149255e-06,
|
| 1522 |
+
"loss": 0.609,
|
| 1523 |
+
"step": 89500
|
| 1524 |
+
},
|
| 1525 |
+
{
|
| 1526 |
+
"epoch": 8.241535608818076,
|
| 1527 |
+
"grad_norm": 2.343564748764038,
|
| 1528 |
+
"learning_rate": 7.177238805970151e-06,
|
| 1529 |
+
"loss": 0.5974,
|
| 1530 |
+
"step": 90000
|
| 1531 |
+
},
|
| 1532 |
+
{
|
| 1533 |
+
"epoch": 8.241535608818076,
|
| 1534 |
+
"eval_loss": 0.24035315215587616,
|
| 1535 |
+
"eval_runtime": 656.8993,
|
| 1536 |
+
"eval_samples_per_second": 9.974,
|
| 1537 |
+
"eval_steps_per_second": 9.974,
|
| 1538 |
+
"eval_wer": 0.11389869877462018,
|
| 1539 |
+
"step": 90000
|
| 1540 |
+
},
|
| 1541 |
+
{
|
| 1542 |
+
"epoch": 8.287320010072568,
|
| 1543 |
+
"grad_norm": 3.118222951889038,
|
| 1544 |
+
"learning_rate": 6.9906716417910455e-06,
|
| 1545 |
+
"loss": 0.6189,
|
| 1546 |
+
"step": 90500
|
| 1547 |
+
},
|
| 1548 |
+
{
|
| 1549 |
+
"epoch": 8.33310441132706,
|
| 1550 |
+
"grad_norm": 1.9298070669174194,
|
| 1551 |
+
"learning_rate": 6.804104477611941e-06,
|
| 1552 |
+
"loss": 0.6321,
|
| 1553 |
+
"step": 91000
|
| 1554 |
+
},
|
| 1555 |
+
{
|
| 1556 |
+
"epoch": 8.378888812581554,
|
| 1557 |
+
"grad_norm": 2.3452179431915283,
|
| 1558 |
+
"learning_rate": 6.617537313432837e-06,
|
| 1559 |
+
"loss": 0.6131,
|
| 1560 |
+
"step": 91500
|
| 1561 |
+
},
|
| 1562 |
+
{
|
| 1563 |
+
"epoch": 8.424673213836046,
|
| 1564 |
+
"grad_norm": 1.260360836982727,
|
| 1565 |
+
"learning_rate": 6.43134328358209e-06,
|
| 1566 |
+
"loss": 0.6485,
|
| 1567 |
+
"step": 92000
|
| 1568 |
+
},
|
| 1569 |
+
{
|
| 1570 |
+
"epoch": 8.470457615090538,
|
| 1571 |
+
"grad_norm": 2.775022506713867,
|
| 1572 |
+
"learning_rate": 6.2447761194029854e-06,
|
| 1573 |
+
"loss": 0.6397,
|
| 1574 |
+
"step": 92500
|
| 1575 |
+
},
|
| 1576 |
+
{
|
| 1577 |
+
"epoch": 8.516242016345032,
|
| 1578 |
+
"grad_norm": 2.084218740463257,
|
| 1579 |
+
"learning_rate": 6.058208955223881e-06,
|
| 1580 |
+
"loss": 0.5994,
|
| 1581 |
+
"step": 93000
|
| 1582 |
+
},
|
| 1583 |
+
{
|
| 1584 |
+
"epoch": 8.516242016345032,
|
| 1585 |
+
"eval_loss": 0.2302207350730896,
|
| 1586 |
+
"eval_runtime": 661.0282,
|
| 1587 |
+
"eval_samples_per_second": 9.912,
|
| 1588 |
+
"eval_steps_per_second": 9.912,
|
| 1589 |
+
"eval_wer": 0.11068118793107534,
|
| 1590 |
+
"step": 93000
|
| 1591 |
+
},
|
| 1592 |
+
{
|
| 1593 |
+
"epoch": 8.562026417599524,
|
| 1594 |
+
"grad_norm": 3.495532751083374,
|
| 1595 |
+
"learning_rate": 5.871641791044776e-06,
|
| 1596 |
+
"loss": 0.6213,
|
| 1597 |
+
"step": 93500
|
| 1598 |
+
},
|
| 1599 |
+
{
|
| 1600 |
+
"epoch": 8.607810818854016,
|
| 1601 |
+
"grad_norm": 1.2543615102767944,
|
| 1602 |
+
"learning_rate": 5.68544776119403e-06,
|
| 1603 |
+
"loss": 0.6118,
|
| 1604 |
+
"step": 94000
|
| 1605 |
+
},
|
| 1606 |
+
{
|
| 1607 |
+
"epoch": 8.653595220108508,
|
| 1608 |
+
"grad_norm": 2.4980568885803223,
|
| 1609 |
+
"learning_rate": 5.498880597014926e-06,
|
| 1610 |
+
"loss": 0.5992,
|
| 1611 |
+
"step": 94500
|
| 1612 |
+
},
|
| 1613 |
+
{
|
| 1614 |
+
"epoch": 8.699379621363002,
|
| 1615 |
+
"grad_norm": 3.1688904762268066,
|
| 1616 |
+
"learning_rate": 5.312313432835822e-06,
|
| 1617 |
+
"loss": 0.629,
|
| 1618 |
+
"step": 95000
|
| 1619 |
+
},
|
| 1620 |
+
{
|
| 1621 |
+
"epoch": 8.745164022617494,
|
| 1622 |
+
"grad_norm": 1.777855396270752,
|
| 1623 |
+
"learning_rate": 5.125746268656717e-06,
|
| 1624 |
+
"loss": 0.5818,
|
| 1625 |
+
"step": 95500
|
| 1626 |
+
},
|
| 1627 |
+
{
|
| 1628 |
+
"epoch": 8.790948423871987,
|
| 1629 |
+
"grad_norm": 4.508159637451172,
|
| 1630 |
+
"learning_rate": 4.939179104477612e-06,
|
| 1631 |
+
"loss": 0.6323,
|
| 1632 |
+
"step": 96000
|
| 1633 |
+
},
|
| 1634 |
+
{
|
| 1635 |
+
"epoch": 8.790948423871987,
|
| 1636 |
+
"eval_loss": 0.23431606590747833,
|
| 1637 |
+
"eval_runtime": 665.4475,
|
| 1638 |
+
"eval_samples_per_second": 9.846,
|
| 1639 |
+
"eval_steps_per_second": 9.846,
|
| 1640 |
+
"eval_wer": 0.11001753928562964,
|
| 1641 |
+
"step": 96000
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 8.836732825126479,
|
| 1645 |
+
"grad_norm": 2.9219791889190674,
|
| 1646 |
+
"learning_rate": 4.752611940298508e-06,
|
| 1647 |
+
"loss": 0.5897,
|
| 1648 |
+
"step": 96500
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 8.882517226380973,
|
| 1652 |
+
"grad_norm": 1.1439307928085327,
|
| 1653 |
+
"learning_rate": 4.5660447761194035e-06,
|
| 1654 |
+
"loss": 0.611,
|
| 1655 |
+
"step": 97000
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 8.928301627635465,
|
| 1659 |
+
"grad_norm": 2.547057628631592,
|
| 1660 |
+
"learning_rate": 4.379477611940298e-06,
|
| 1661 |
+
"loss": 0.6163,
|
| 1662 |
+
"step": 97500
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"epoch": 8.974086028889957,
|
| 1666 |
+
"grad_norm": 1.996872901916504,
|
| 1667 |
+
"learning_rate": 4.193283582089553e-06,
|
| 1668 |
+
"loss": 0.6063,
|
| 1669 |
+
"step": 98000
|
| 1670 |
+
},
|
| 1671 |
+
{
|
| 1672 |
+
"epoch": 9.019870430144449,
|
| 1673 |
+
"grad_norm": 1.6827927827835083,
|
| 1674 |
+
"learning_rate": 4.006716417910448e-06,
|
| 1675 |
+
"loss": 0.594,
|
| 1676 |
+
"step": 98500
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 9.065654831398943,
|
| 1680 |
+
"grad_norm": 1.5130544900894165,
|
| 1681 |
+
"learning_rate": 3.820149253731343e-06,
|
| 1682 |
+
"loss": 0.5862,
|
| 1683 |
+
"step": 99000
|
| 1684 |
+
},
|
| 1685 |
+
{
|
| 1686 |
+
"epoch": 9.065654831398943,
|
| 1687 |
+
"eval_loss": 0.22885525226593018,
|
| 1688 |
+
"eval_runtime": 661.9197,
|
| 1689 |
+
"eval_samples_per_second": 9.898,
|
| 1690 |
+
"eval_steps_per_second": 9.898,
|
| 1691 |
+
"eval_wer": 0.10877912350975326,
|
| 1692 |
+
"step": 99000
|
| 1693 |
+
},
|
| 1694 |
+
{
|
| 1695 |
+
"epoch": 9.111439232653435,
|
| 1696 |
+
"grad_norm": 2.690355062484741,
|
| 1697 |
+
"learning_rate": 3.6335820895522388e-06,
|
| 1698 |
+
"loss": 0.5986,
|
| 1699 |
+
"step": 99500
|
| 1700 |
+
},
|
| 1701 |
+
{
|
| 1702 |
+
"epoch": 9.157223633907927,
|
| 1703 |
+
"grad_norm": 1.3742425441741943,
|
| 1704 |
+
"learning_rate": 3.4470149253731346e-06,
|
| 1705 |
+
"loss": 0.5974,
|
| 1706 |
+
"step": 100000
|
| 1707 |
+
},
|
| 1708 |
+
{
|
| 1709 |
+
"epoch": 9.20300803516242,
|
| 1710 |
+
"grad_norm": 2.864790678024292,
|
| 1711 |
+
"learning_rate": 3.2608208955223884e-06,
|
| 1712 |
+
"loss": 0.5916,
|
| 1713 |
+
"step": 100500
|
| 1714 |
+
},
|
| 1715 |
+
{
|
| 1716 |
+
"epoch": 9.248792436416913,
|
| 1717 |
+
"grad_norm": 1.65906822681427,
|
| 1718 |
+
"learning_rate": 3.074253731343284e-06,
|
| 1719 |
+
"loss": 0.5724,
|
| 1720 |
+
"step": 101000
|
| 1721 |
+
},
|
| 1722 |
+
{
|
| 1723 |
+
"epoch": 9.294576837671405,
|
| 1724 |
+
"grad_norm": 1.9599213600158691,
|
| 1725 |
+
"learning_rate": 2.8876865671641795e-06,
|
| 1726 |
+
"loss": 0.5857,
|
| 1727 |
+
"step": 101500
|
| 1728 |
+
},
|
| 1729 |
+
{
|
| 1730 |
+
"epoch": 9.340361238925897,
|
| 1731 |
+
"grad_norm": 3.1552865505218506,
|
| 1732 |
+
"learning_rate": 2.7011194029850745e-06,
|
| 1733 |
+
"loss": 0.6196,
|
| 1734 |
+
"step": 102000
|
| 1735 |
+
},
|
| 1736 |
+
{
|
| 1737 |
+
"epoch": 9.340361238925897,
|
| 1738 |
+
"eval_loss": 0.23824112117290497,
|
| 1739 |
+
"eval_runtime": 661.8149,
|
| 1740 |
+
"eval_samples_per_second": 9.9,
|
| 1741 |
+
"eval_steps_per_second": 9.9,
|
| 1742 |
+
"eval_wer": 0.10700149320945225,
|
| 1743 |
+
"step": 102000
|
| 1744 |
+
},
|
| 1745 |
+
{
|
| 1746 |
+
"epoch": 9.386145640180391,
|
| 1747 |
+
"grad_norm": 2.2402780055999756,
|
| 1748 |
+
"learning_rate": 2.5145522388059703e-06,
|
| 1749 |
+
"loss": 0.6165,
|
| 1750 |
+
"step": 102500
|
| 1751 |
+
},
|
| 1752 |
+
{
|
| 1753 |
+
"epoch": 9.431930041434883,
|
| 1754 |
+
"grad_norm": 3.58760929107666,
|
| 1755 |
+
"learning_rate": 2.327985074626866e-06,
|
| 1756 |
+
"loss": 0.584,
|
| 1757 |
+
"step": 103000
|
| 1758 |
+
},
|
| 1759 |
+
{
|
| 1760 |
+
"epoch": 9.477714442689376,
|
| 1761 |
+
"grad_norm": 4.692273139953613,
|
| 1762 |
+
"learning_rate": 2.1417910447761194e-06,
|
| 1763 |
+
"loss": 0.5801,
|
| 1764 |
+
"step": 103500
|
| 1765 |
+
},
|
| 1766 |
+
{
|
| 1767 |
+
"epoch": 9.523498843943868,
|
| 1768 |
+
"grad_norm": 1.1370134353637695,
|
| 1769 |
+
"learning_rate": 1.955223880597015e-06,
|
| 1770 |
+
"loss": 0.5876,
|
| 1771 |
+
"step": 104000
|
| 1772 |
+
},
|
| 1773 |
+
{
|
| 1774 |
+
"epoch": 9.569283245198362,
|
| 1775 |
+
"grad_norm": 1.0059013366699219,
|
| 1776 |
+
"learning_rate": 1.7686567164179106e-06,
|
| 1777 |
+
"loss": 0.6329,
|
| 1778 |
+
"step": 104500
|
| 1779 |
+
},
|
| 1780 |
+
{
|
| 1781 |
+
"epoch": 9.615067646452854,
|
| 1782 |
+
"grad_norm": 2.666593551635742,
|
| 1783 |
+
"learning_rate": 1.582089552238806e-06,
|
| 1784 |
+
"loss": 0.6077,
|
| 1785 |
+
"step": 105000
|
| 1786 |
+
},
|
| 1787 |
+
{
|
| 1788 |
+
"epoch": 9.615067646452854,
|
| 1789 |
+
"eval_loss": 0.22313112020492554,
|
| 1790 |
+
"eval_runtime": 661.0901,
|
| 1791 |
+
"eval_samples_per_second": 9.911,
|
| 1792 |
+
"eval_steps_per_second": 9.911,
|
| 1793 |
+
"eval_wer": 0.10743997535019317,
|
| 1794 |
+
"step": 105000
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 9.660852047707346,
|
| 1798 |
+
"grad_norm": 3.046264410018921,
|
| 1799 |
+
"learning_rate": 1.3955223880597016e-06,
|
| 1800 |
+
"loss": 0.5714,
|
| 1801 |
+
"step": 105500
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 9.706636448961838,
|
| 1805 |
+
"grad_norm": 1.997536301612854,
|
| 1806 |
+
"learning_rate": 1.2089552238805971e-06,
|
| 1807 |
+
"loss": 0.6201,
|
| 1808 |
+
"step": 106000
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 9.752420850216332,
|
| 1812 |
+
"grad_norm": 3.185781478881836,
|
| 1813 |
+
"learning_rate": 1.0223880597014927e-06,
|
| 1814 |
+
"loss": 0.5782,
|
| 1815 |
+
"step": 106500
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 9.798205251470824,
|
| 1819 |
+
"grad_norm": 4.050471782684326,
|
| 1820 |
+
"learning_rate": 8.361940298507463e-07,
|
| 1821 |
+
"loss": 0.5841,
|
| 1822 |
+
"step": 107000
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 9.843989652725316,
|
| 1826 |
+
"grad_norm": 2.3069846630096436,
|
| 1827 |
+
"learning_rate": 6.496268656716419e-07,
|
| 1828 |
+
"loss": 0.5601,
|
| 1829 |
+
"step": 107500
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 9.88977405397981,
|
| 1833 |
+
"grad_norm": 3.8783347606658936,
|
| 1834 |
+
"learning_rate": 4.6343283582089555e-07,
|
| 1835 |
+
"loss": 0.5949,
|
| 1836 |
+
"step": 108000
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 9.88977405397981,
|
| 1840 |
+
"eval_loss": 0.22683313488960266,
|
| 1841 |
+
"eval_runtime": 661.0107,
|
| 1842 |
+
"eval_samples_per_second": 9.912,
|
| 1843 |
+
"eval_steps_per_second": 9.912,
|
| 1844 |
+
"eval_wer": 0.1066222654120547,
|
| 1845 |
+
"step": 108000
|
| 1846 |
+
},
|
| 1847 |
+
{
|
| 1848 |
+
"epoch": 9.935558455234302,
|
| 1849 |
+
"grad_norm": 1.1555778980255127,
|
| 1850 |
+
"learning_rate": 2.768656716417911e-07,
|
| 1851 |
+
"loss": 0.5902,
|
| 1852 |
+
"step": 108500
|
| 1853 |
+
},
|
| 1854 |
+
{
|
| 1855 |
+
"epoch": 9.981342856488794,
|
| 1856 |
+
"grad_norm": 0.8959473371505737,
|
| 1857 |
+
"learning_rate": 9.029850746268658e-08,
|
| 1858 |
+
"loss": 0.5785,
|
| 1859 |
+
"step": 109000
|
| 1860 |
+
}
|
| 1861 |
+
],
|
| 1862 |
+
"logging_steps": 500,
|
| 1863 |
+
"max_steps": 109200,
|
| 1864 |
+
"num_input_tokens_seen": 0,
|
| 1865 |
+
"num_train_epochs": 10,
|
| 1866 |
+
"save_steps": 3000,
|
| 1867 |
+
"stateful_callbacks": {
|
| 1868 |
+
"TrainerControl": {
|
| 1869 |
+
"args": {
|
| 1870 |
+
"should_epoch_stop": false,
|
| 1871 |
+
"should_evaluate": false,
|
| 1872 |
+
"should_log": false,
|
| 1873 |
+
"should_save": true,
|
| 1874 |
+
"should_training_stop": true
|
| 1875 |
+
},
|
| 1876 |
+
"attributes": {}
|
| 1877 |
+
}
|
| 1878 |
+
},
|
| 1879 |
+
"total_flos": 1.0560519550521973e+20,
|
| 1880 |
+
"train_batch_size": 1,
|
| 1881 |
+
"trial_name": null,
|
| 1882 |
+
"trial_params": null
|
| 1883 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fefce5260df9a154d9fd9f7f8daf320142bab60fafb944da85851dc8bc8418a0
|
| 3 |
+
size 5432
|
vocab.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"ਂ": 1, "ਅ": 2, "ਆ": 3, "ਇ": 4, "ਈ": 5, "ਉ": 6, "ਊ": 7, "ਏ": 8, "ਐ": 9, "ਓ": 10, "ਔ": 11, "ਕ": 12, "ਖ": 13, "ਗ": 14, "ਘ": 15, "ਚ": 16, "ਛ": 17, "ਜ": 18, "ਝ": 19, "ਟ": 20, "ਠ": 21, "ਡ": 22, "ਢ": 23, "ਣ": 24, "ਤ": 25, "ਥ": 26, "ਦ": 27, "ਧ": 28, "ਨ": 29, "ਪ": 30, "ਫ": 31, "ਬ": 32, "ਭ": 33, "ਮ": 34, "ਯ": 35, "ਰ": 36, "ਲ": 37, "ਲ਼": 38, "ਵ": 39, "ਸ਼": 40, "ਸ": 41, "ਹ": 42, "਼": 43, "ਾ": 44, "ਿ": 45, "ੀ": 46, "ੁ": 47, "ੂ": 48, "ੇ": 49, "ੈ": 50, "ੋ": 51, "ੌ": 52, "੍": 53, "ਖ਼": 54, "ਗ਼": 55, "ਜ਼": 56, "ੜ": 57, "ਫ਼": 58, "ੰ": 59, "ੱ": 60, "ੲ": 61, "ੳ": 62, "|": 0, "[UNK]": 63, "[PAD]": 64}
|