Upload folder using huggingface_hub
Browse files- README_new.md +69 -0
- all_results.json +13 -13
- config.json +3 -6
- eval_results.json +7 -7
- generation_config.json +2 -2
- model-00001-of-00002.safetensors +1 -1
- model-00002-of-00002.safetensors +1 -1
- runs/Mar22_10-30-53_d4a283a05c40/events.out.tfevents.1742640704.d4a283a05c40.213.0 +3 -0
- runs/Mar22_10-30-53_d4a283a05c40/events.out.tfevents.1742706967.d4a283a05c40.213.1 +3 -0
- tokenizer.json +0 -0
- tokenizer_config.json +1 -0
- train_results.json +7 -7
- trainer_state.json +741 -741
- training_args.bin +2 -2
README_new.md
ADDED
|
@@ -0,0 +1,69 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: transformers
|
| 3 |
+
license: apache-2.0
|
| 4 |
+
base_model: openai/whisper-large-v3
|
| 5 |
+
tags:
|
| 6 |
+
- generated_from_trainer
|
| 7 |
+
metrics:
|
| 8 |
+
- wer
|
| 9 |
+
model-index:
|
| 10 |
+
- name: whisper-large-v3-ft-btb-cv-cvad-ca-cy-2503
|
| 11 |
+
results: []
|
| 12 |
+
---
|
| 13 |
+
|
| 14 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 15 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 16 |
+
|
| 17 |
+
# whisper-large-v3-ft-btb-cv-cvad-ca-cy-2503
|
| 18 |
+
|
| 19 |
+
This model is a fine-tuned version of [openai/whisper-large-v3](https://huggingface.co/openai/whisper-large-v3) on the DewiBrynJones/banc-trawsgrifiadau-bangor train main, DewiBrynJones/commonvoice_18_0_cy train+dev+other_with_excluded main, cymen-arfor/lleisiau-arfor train+dev main, techiaith/commonvoice_vad_cy train main dataset.
|
| 20 |
+
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 0.3739
|
| 22 |
+
- Wer: 0.2915
|
| 23 |
+
|
| 24 |
+
## Model description
|
| 25 |
+
|
| 26 |
+
More information needed
|
| 27 |
+
|
| 28 |
+
## Intended uses & limitations
|
| 29 |
+
|
| 30 |
+
More information needed
|
| 31 |
+
|
| 32 |
+
## Training and evaluation data
|
| 33 |
+
|
| 34 |
+
More information needed
|
| 35 |
+
|
| 36 |
+
## Training procedure
|
| 37 |
+
|
| 38 |
+
### Training hyperparameters
|
| 39 |
+
|
| 40 |
+
The following hyperparameters were used during training:
|
| 41 |
+
- learning_rate: 1e-05
|
| 42 |
+
- train_batch_size: 16
|
| 43 |
+
- eval_batch_size: 16
|
| 44 |
+
- seed: 42
|
| 45 |
+
- gradient_accumulation_steps: 2
|
| 46 |
+
- total_train_batch_size: 32
|
| 47 |
+
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
|
| 48 |
+
- lr_scheduler_type: linear
|
| 49 |
+
- lr_scheduler_warmup_steps: 500
|
| 50 |
+
- training_steps: 5000
|
| 51 |
+
- mixed_precision_training: Native AMP
|
| 52 |
+
|
| 53 |
+
### Training results
|
| 54 |
+
|
| 55 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
| 56 |
+
|:-------------:|:------:|:----:|:---------------:|:------:|
|
| 57 |
+
| 0.5304 | 0.3240 | 1000 | 0.5236 | 0.3786 |
|
| 58 |
+
| 0.4409 | 0.6480 | 2000 | 0.4458 | 0.3563 |
|
| 59 |
+
| 0.3914 | 0.9720 | 3000 | 0.4035 | 0.3090 |
|
| 60 |
+
| 0.296 | 1.2958 | 4000 | 0.3868 | 0.2977 |
|
| 61 |
+
| 0.274 | 1.6198 | 5000 | 0.3739 | 0.2915 |
|
| 62 |
+
|
| 63 |
+
|
| 64 |
+
### Framework versions
|
| 65 |
+
|
| 66 |
+
- Transformers 4.49.0
|
| 67 |
+
- Pytorch 2.6.0+cu124
|
| 68 |
+
- Datasets 3.3.2
|
| 69 |
+
- Tokenizers 0.21.0
|
all_results.json
CHANGED
|
@@ -1,15 +1,15 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_loss": 0.
|
| 4 |
-
"eval_runtime":
|
| 5 |
-
"eval_samples":
|
| 6 |
-
"eval_samples_per_second": 2.
|
| 7 |
-
"eval_steps_per_second": 0.
|
| 8 |
-
"eval_wer": 0.
|
| 9 |
-
"total_flos": 5.
|
| 10 |
-
"train_loss": 0.
|
| 11 |
-
"train_runtime":
|
| 12 |
-
"train_samples":
|
| 13 |
-
"train_samples_per_second": 2.
|
| 14 |
-
"train_steps_per_second": 0.
|
| 15 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 1.619795885306982,
|
| 3 |
+
"eval_loss": 0.3738669753074646,
|
| 4 |
+
"eval_runtime": 1862.9517,
|
| 5 |
+
"eval_samples": 3897,
|
| 6 |
+
"eval_samples_per_second": 2.092,
|
| 7 |
+
"eval_steps_per_second": 0.131,
|
| 8 |
+
"eval_wer": 0.29154381412902464,
|
| 9 |
+
"total_flos": 5.435419715783885e+20,
|
| 10 |
+
"train_loss": 0.4225531764030456,
|
| 11 |
+
"train_runtime": 64153.0147,
|
| 12 |
+
"train_samples": 98767,
|
| 13 |
+
"train_samples_per_second": 2.494,
|
| 14 |
+
"train_steps_per_second": 0.078
|
| 15 |
}
|
config.json
CHANGED
|
@@ -7,10 +7,7 @@
|
|
| 7 |
"WhisperForConditionalGeneration"
|
| 8 |
],
|
| 9 |
"attention_dropout": 0.0,
|
| 10 |
-
"begin_suppress_tokens":
|
| 11 |
-
220,
|
| 12 |
-
50257
|
| 13 |
-
],
|
| 14 |
"bos_token_id": 50257,
|
| 15 |
"classifier_proj_size": 256,
|
| 16 |
"d_model": 1280,
|
|
@@ -34,7 +31,7 @@
|
|
| 34 |
"mask_time_length": 10,
|
| 35 |
"mask_time_min_masks": 2,
|
| 36 |
"mask_time_prob": 0.05,
|
| 37 |
-
"max_length":
|
| 38 |
"max_source_positions": 1500,
|
| 39 |
"max_target_positions": 448,
|
| 40 |
"median_filter_width": 7,
|
|
@@ -44,7 +41,7 @@
|
|
| 44 |
"pad_token_id": 50256,
|
| 45 |
"scale_embedding": false,
|
| 46 |
"torch_dtype": "float32",
|
| 47 |
-
"transformers_version": "4.
|
| 48 |
"use_cache": true,
|
| 49 |
"use_weighted_layer_sum": false,
|
| 50 |
"vocab_size": 51866
|
|
|
|
| 7 |
"WhisperForConditionalGeneration"
|
| 8 |
],
|
| 9 |
"attention_dropout": 0.0,
|
| 10 |
+
"begin_suppress_tokens": null,
|
|
|
|
|
|
|
|
|
|
| 11 |
"bos_token_id": 50257,
|
| 12 |
"classifier_proj_size": 256,
|
| 13 |
"d_model": 1280,
|
|
|
|
| 31 |
"mask_time_length": 10,
|
| 32 |
"mask_time_min_masks": 2,
|
| 33 |
"mask_time_prob": 0.05,
|
| 34 |
+
"max_length": null,
|
| 35 |
"max_source_positions": 1500,
|
| 36 |
"max_target_positions": 448,
|
| 37 |
"median_filter_width": 7,
|
|
|
|
| 41 |
"pad_token_id": 50256,
|
| 42 |
"scale_embedding": false,
|
| 43 |
"torch_dtype": "float32",
|
| 44 |
+
"transformers_version": "4.49.0",
|
| 45 |
"use_cache": true,
|
| 46 |
"use_weighted_layer_sum": false,
|
| 47 |
"vocab_size": 51866
|
eval_results.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_loss": 0.
|
| 4 |
-
"eval_runtime":
|
| 5 |
-
"eval_samples":
|
| 6 |
-
"eval_samples_per_second": 2.
|
| 7 |
-
"eval_steps_per_second": 0.
|
| 8 |
-
"eval_wer": 0.
|
| 9 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 1.619795885306982,
|
| 3 |
+
"eval_loss": 0.3738669753074646,
|
| 4 |
+
"eval_runtime": 1862.9517,
|
| 5 |
+
"eval_samples": 3897,
|
| 6 |
+
"eval_samples_per_second": 2.092,
|
| 7 |
+
"eval_steps_per_second": 0.131,
|
| 8 |
+
"eval_wer": 0.29154381412902464
|
| 9 |
}
|
generation_config.json
CHANGED
|
@@ -151,7 +151,7 @@
|
|
| 151 |
"<|yue|>": 50358,
|
| 152 |
"<|zh|>": 50260
|
| 153 |
},
|
| 154 |
-
"language":
|
| 155 |
"max_initial_timestamp_index": 50,
|
| 156 |
"max_length": 448,
|
| 157 |
"no_timestamps_token_id": 50364,
|
|
@@ -253,5 +253,5 @@
|
|
| 253 |
"transcribe": 50360,
|
| 254 |
"translate": 50359
|
| 255 |
},
|
| 256 |
-
"transformers_version": "4.
|
| 257 |
}
|
|
|
|
| 151 |
"<|yue|>": 50358,
|
| 152 |
"<|zh|>": 50260
|
| 153 |
},
|
| 154 |
+
"language": null,
|
| 155 |
"max_initial_timestamp_index": 50,
|
| 156 |
"max_length": 448,
|
| 157 |
"no_timestamps_token_id": 50364,
|
|
|
|
| 253 |
"transcribe": 50360,
|
| 254 |
"translate": 50359
|
| 255 |
},
|
| 256 |
+
"transformers_version": "4.49.0"
|
| 257 |
}
|
model-00001-of-00002.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4993448880
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:89026adf241ab6e6f7d62cc3a2c3e4007c8b34ef54d1222ee9d8e204ccb4653d
|
| 3 |
size 4993448880
|
model-00002-of-00002.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1180663192
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d5eb80d43bb52300a1166a1892b4d9fa06d3b0546ffb4338f7caa0422698d839
|
| 3 |
size 1180663192
|
runs/Mar22_10-30-53_d4a283a05c40/events.out.tfevents.1742640704.d4a283a05c40.213.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:da50f05716832fd4988146bdb5ba728d3f7382fe0cefec5700063559e37a0e67
|
| 3 |
+
size 50135
|
runs/Mar22_10-30-53_d4a283a05c40/events.out.tfevents.1742706967.d4a283a05c40.213.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:054c3fa54b2f8c098eb2a448fa6f33487c4a63baa3d8f235cbef0fa6d33f7fbc
|
| 3 |
+
size 406
|
tokenizer.json
CHANGED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
CHANGED
|
@@ -12987,6 +12987,7 @@
|
|
| 12987 |
"clean_up_tokenization_spaces": true,
|
| 12988 |
"eos_token": "<|endoftext|>",
|
| 12989 |
"errors": "replace",
|
|
|
|
| 12990 |
"model_max_length": 1000000000000000019884624838656,
|
| 12991 |
"pad_token": "<|endoftext|>",
|
| 12992 |
"processor_class": "WhisperProcessor",
|
|
|
|
| 12987 |
"clean_up_tokenization_spaces": true,
|
| 12988 |
"eos_token": "<|endoftext|>",
|
| 12989 |
"errors": "replace",
|
| 12990 |
+
"extra_special_tokens": {},
|
| 12991 |
"model_max_length": 1000000000000000019884624838656,
|
| 12992 |
"pad_token": "<|endoftext|>",
|
| 12993 |
"processor_class": "WhisperProcessor",
|
train_results.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"total_flos": 5.
|
| 4 |
-
"train_loss": 0.
|
| 5 |
-
"train_runtime":
|
| 6 |
-
"train_samples":
|
| 7 |
-
"train_samples_per_second": 2.
|
| 8 |
-
"train_steps_per_second": 0.
|
| 9 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 1.619795885306982,
|
| 3 |
+
"total_flos": 5.435419715783885e+20,
|
| 4 |
+
"train_loss": 0.4225531764030456,
|
| 5 |
+
"train_runtime": 64153.0147,
|
| 6 |
+
"train_samples": 98767,
|
| 7 |
+
"train_samples_per_second": 2.494,
|
| 8 |
+
"train_steps_per_second": 0.078
|
| 9 |
}
|
trainer_state.json
CHANGED
|
@@ -1,7 +1,7 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch":
|
| 5 |
"eval_steps": 1000,
|
| 6 |
"global_step": 5000,
|
| 7 |
"is_hyper_param_search": false,
|
|
@@ -9,1464 +9,1464 @@
|
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
-
"epoch": 0.
|
| 13 |
-
"grad_norm":
|
| 14 |
"learning_rate": 5.000000000000001e-07,
|
| 15 |
-
"loss": 1.
|
| 16 |
"step": 25
|
| 17 |
},
|
| 18 |
{
|
| 19 |
-
"epoch": 0.
|
| 20 |
-
"grad_norm": 5.
|
| 21 |
"learning_rate": 1.0000000000000002e-06,
|
| 22 |
-
"loss": 1.
|
| 23 |
"step": 50
|
| 24 |
},
|
| 25 |
{
|
| 26 |
-
"epoch": 0.
|
| 27 |
-
"grad_norm":
|
| 28 |
"learning_rate": 1.5e-06,
|
| 29 |
-
"loss":
|
| 30 |
"step": 75
|
| 31 |
},
|
| 32 |
{
|
| 33 |
-
"epoch": 0.
|
| 34 |
-
"grad_norm": 4.
|
| 35 |
"learning_rate": 2.0000000000000003e-06,
|
| 36 |
-
"loss": 0.
|
| 37 |
"step": 100
|
| 38 |
},
|
| 39 |
{
|
| 40 |
-
"epoch": 0.
|
| 41 |
-
"grad_norm":
|
| 42 |
"learning_rate": 2.5e-06,
|
| 43 |
-
"loss": 0.
|
| 44 |
"step": 125
|
| 45 |
},
|
| 46 |
{
|
| 47 |
-
"epoch": 0.
|
| 48 |
-
"grad_norm":
|
| 49 |
"learning_rate": 3e-06,
|
| 50 |
-
"loss": 0.
|
| 51 |
"step": 150
|
| 52 |
},
|
| 53 |
{
|
| 54 |
-
"epoch": 0.
|
| 55 |
-
"grad_norm":
|
| 56 |
"learning_rate": 3.5e-06,
|
| 57 |
-
"loss": 0.
|
| 58 |
"step": 175
|
| 59 |
},
|
| 60 |
{
|
| 61 |
-
"epoch": 0.
|
| 62 |
-
"grad_norm": 5.
|
| 63 |
"learning_rate": 4.000000000000001e-06,
|
| 64 |
-
"loss": 0.
|
| 65 |
"step": 200
|
| 66 |
},
|
| 67 |
{
|
| 68 |
-
"epoch": 0.
|
| 69 |
-
"grad_norm":
|
| 70 |
"learning_rate": 4.5e-06,
|
| 71 |
-
"loss": 0.
|
| 72 |
"step": 225
|
| 73 |
},
|
| 74 |
{
|
| 75 |
-
"epoch": 0.
|
| 76 |
-
"grad_norm":
|
| 77 |
"learning_rate": 5e-06,
|
| 78 |
-
"loss": 0.
|
| 79 |
"step": 250
|
| 80 |
},
|
| 81 |
{
|
| 82 |
-
"epoch": 0.
|
| 83 |
-
"grad_norm":
|
| 84 |
"learning_rate": 5.500000000000001e-06,
|
| 85 |
-
"loss": 0.
|
| 86 |
"step": 275
|
| 87 |
},
|
| 88 |
{
|
| 89 |
-
"epoch": 0.
|
| 90 |
-
"grad_norm": 4.
|
| 91 |
"learning_rate": 6e-06,
|
| 92 |
-
"loss": 0.
|
| 93 |
"step": 300
|
| 94 |
},
|
| 95 |
{
|
| 96 |
-
"epoch": 0.
|
| 97 |
-
"grad_norm": 4.
|
| 98 |
"learning_rate": 6.5000000000000004e-06,
|
| 99 |
-
"loss": 0.
|
| 100 |
"step": 325
|
| 101 |
},
|
| 102 |
{
|
| 103 |
-
"epoch": 0.
|
| 104 |
-
"grad_norm":
|
| 105 |
"learning_rate": 7e-06,
|
| 106 |
-
"loss": 0.
|
| 107 |
"step": 350
|
| 108 |
},
|
| 109 |
{
|
| 110 |
-
"epoch": 0.
|
| 111 |
-
"grad_norm":
|
| 112 |
"learning_rate": 7.500000000000001e-06,
|
| 113 |
-
"loss": 0.
|
| 114 |
"step": 375
|
| 115 |
},
|
| 116 |
{
|
| 117 |
-
"epoch": 0.
|
| 118 |
-
"grad_norm":
|
| 119 |
"learning_rate": 8.000000000000001e-06,
|
| 120 |
-
"loss": 0.
|
| 121 |
"step": 400
|
| 122 |
},
|
| 123 |
{
|
| 124 |
-
"epoch": 0.
|
| 125 |
-
"grad_norm":
|
| 126 |
"learning_rate": 8.5e-06,
|
| 127 |
-
"loss": 0.
|
| 128 |
"step": 425
|
| 129 |
},
|
| 130 |
{
|
| 131 |
-
"epoch": 0.
|
| 132 |
-
"grad_norm": 4.
|
| 133 |
"learning_rate": 9e-06,
|
| 134 |
-
"loss": 0.
|
| 135 |
"step": 450
|
| 136 |
},
|
| 137 |
{
|
| 138 |
-
"epoch": 0.
|
| 139 |
-
"grad_norm":
|
| 140 |
"learning_rate": 9.5e-06,
|
| 141 |
-
"loss": 0.
|
| 142 |
"step": 475
|
| 143 |
},
|
| 144 |
{
|
| 145 |
-
"epoch": 0.
|
| 146 |
-
"grad_norm":
|
| 147 |
"learning_rate": 1e-05,
|
| 148 |
-
"loss": 0.
|
| 149 |
"step": 500
|
| 150 |
},
|
| 151 |
{
|
| 152 |
-
"epoch": 0.
|
| 153 |
-
"grad_norm":
|
| 154 |
"learning_rate": 9.944444444444445e-06,
|
| 155 |
-
"loss": 0.
|
| 156 |
"step": 525
|
| 157 |
},
|
| 158 |
{
|
| 159 |
-
"epoch": 0.
|
| 160 |
-
"grad_norm":
|
| 161 |
"learning_rate": 9.88888888888889e-06,
|
| 162 |
-
"loss": 0.
|
| 163 |
"step": 550
|
| 164 |
},
|
| 165 |
{
|
| 166 |
-
"epoch": 0.
|
| 167 |
-
"grad_norm": 3.
|
| 168 |
"learning_rate": 9.833333333333333e-06,
|
| 169 |
-
"loss": 0.
|
| 170 |
"step": 575
|
| 171 |
},
|
| 172 |
{
|
| 173 |
-
"epoch": 0.
|
| 174 |
-
"grad_norm": 3.
|
| 175 |
"learning_rate": 9.777777777777779e-06,
|
| 176 |
-
"loss": 0.
|
| 177 |
"step": 600
|
| 178 |
},
|
| 179 |
{
|
| 180 |
-
"epoch": 0.
|
| 181 |
-
"grad_norm":
|
| 182 |
"learning_rate": 9.722222222222223e-06,
|
| 183 |
-
"loss": 0.
|
| 184 |
"step": 625
|
| 185 |
},
|
| 186 |
{
|
| 187 |
-
"epoch": 0.
|
| 188 |
-
"grad_norm":
|
| 189 |
"learning_rate": 9.666666666666667e-06,
|
| 190 |
-
"loss": 0.
|
| 191 |
"step": 650
|
| 192 |
},
|
| 193 |
{
|
| 194 |
-
"epoch": 0.
|
| 195 |
-
"grad_norm": 3.
|
| 196 |
"learning_rate": 9.611111111111112e-06,
|
| 197 |
-
"loss": 0.
|
| 198 |
"step": 675
|
| 199 |
},
|
| 200 |
{
|
| 201 |
-
"epoch": 0.
|
| 202 |
-
"grad_norm":
|
| 203 |
"learning_rate": 9.555555555555556e-06,
|
| 204 |
-
"loss": 0.
|
| 205 |
"step": 700
|
| 206 |
},
|
| 207 |
{
|
| 208 |
-
"epoch": 0.
|
| 209 |
-
"grad_norm":
|
| 210 |
"learning_rate": 9.5e-06,
|
| 211 |
-
"loss": 0.
|
| 212 |
"step": 725
|
| 213 |
},
|
| 214 |
{
|
| 215 |
-
"epoch": 0.
|
| 216 |
-
"grad_norm": 3.
|
| 217 |
"learning_rate": 9.444444444444445e-06,
|
| 218 |
-
"loss": 0.
|
| 219 |
"step": 750
|
| 220 |
},
|
| 221 |
{
|
| 222 |
-
"epoch": 0.
|
| 223 |
-
"grad_norm":
|
| 224 |
"learning_rate": 9.38888888888889e-06,
|
| 225 |
-
"loss": 0.
|
| 226 |
"step": 775
|
| 227 |
},
|
| 228 |
{
|
| 229 |
-
"epoch": 0.
|
| 230 |
-
"grad_norm":
|
| 231 |
"learning_rate": 9.333333333333334e-06,
|
| 232 |
-
"loss": 0.
|
| 233 |
"step": 800
|
| 234 |
},
|
| 235 |
{
|
| 236 |
-
"epoch": 0.
|
| 237 |
-
"grad_norm":
|
| 238 |
"learning_rate": 9.277777777777778e-06,
|
| 239 |
-
"loss": 0.
|
| 240 |
"step": 825
|
| 241 |
},
|
| 242 |
{
|
| 243 |
-
"epoch": 0.
|
| 244 |
-
"grad_norm":
|
| 245 |
"learning_rate": 9.222222222222224e-06,
|
| 246 |
-
"loss": 0.
|
| 247 |
"step": 850
|
| 248 |
},
|
| 249 |
{
|
| 250 |
-
"epoch": 0.
|
| 251 |
-
"grad_norm": 3.
|
| 252 |
"learning_rate": 9.166666666666666e-06,
|
| 253 |
-
"loss": 0.
|
| 254 |
"step": 875
|
| 255 |
},
|
| 256 |
{
|
| 257 |
-
"epoch": 0.
|
| 258 |
-
"grad_norm": 3.
|
| 259 |
"learning_rate": 9.111111111111112e-06,
|
| 260 |
-
"loss": 0.
|
| 261 |
"step": 900
|
| 262 |
},
|
| 263 |
{
|
| 264 |
-
"epoch": 0.
|
| 265 |
-
"grad_norm":
|
| 266 |
"learning_rate": 9.055555555555556e-06,
|
| 267 |
-
"loss": 0.
|
| 268 |
"step": 925
|
| 269 |
},
|
| 270 |
{
|
| 271 |
-
"epoch": 0.
|
| 272 |
-
"grad_norm":
|
| 273 |
"learning_rate": 9e-06,
|
| 274 |
-
"loss": 0.
|
| 275 |
"step": 950
|
| 276 |
},
|
| 277 |
{
|
| 278 |
-
"epoch": 0.
|
| 279 |
-
"grad_norm": 3.
|
| 280 |
"learning_rate": 8.944444444444446e-06,
|
| 281 |
-
"loss": 0.
|
| 282 |
"step": 975
|
| 283 |
},
|
| 284 |
{
|
| 285 |
-
"epoch": 0.
|
| 286 |
-
"grad_norm":
|
| 287 |
"learning_rate": 8.888888888888888e-06,
|
| 288 |
-
"loss": 0.
|
| 289 |
"step": 1000
|
| 290 |
},
|
| 291 |
{
|
| 292 |
-
"epoch": 0.
|
| 293 |
-
"eval_loss": 0.
|
| 294 |
-
"eval_runtime":
|
| 295 |
-
"eval_samples_per_second":
|
| 296 |
-
"eval_steps_per_second": 0.
|
| 297 |
-
"eval_wer": 0.
|
| 298 |
"step": 1000
|
| 299 |
},
|
| 300 |
{
|
| 301 |
-
"epoch": 0.
|
| 302 |
-
"grad_norm": 3.
|
| 303 |
"learning_rate": 8.833333333333334e-06,
|
| 304 |
-
"loss": 0.
|
| 305 |
"step": 1025
|
| 306 |
},
|
| 307 |
{
|
| 308 |
-
"epoch": 0.
|
| 309 |
-
"grad_norm": 3.
|
| 310 |
"learning_rate": 8.777777777777778e-06,
|
| 311 |
-
"loss": 0.
|
| 312 |
"step": 1050
|
| 313 |
},
|
| 314 |
{
|
| 315 |
-
"epoch": 0.
|
| 316 |
-
"grad_norm":
|
| 317 |
"learning_rate": 8.722222222222224e-06,
|
| 318 |
-
"loss": 0.
|
| 319 |
"step": 1075
|
| 320 |
},
|
| 321 |
{
|
| 322 |
-
"epoch": 0.
|
| 323 |
-
"grad_norm":
|
| 324 |
"learning_rate": 8.666666666666668e-06,
|
| 325 |
-
"loss": 0.
|
| 326 |
"step": 1100
|
| 327 |
},
|
| 328 |
{
|
| 329 |
-
"epoch": 0.
|
| 330 |
-
"grad_norm":
|
| 331 |
"learning_rate": 8.611111111111112e-06,
|
| 332 |
-
"loss": 0.
|
| 333 |
"step": 1125
|
| 334 |
},
|
| 335 |
{
|
| 336 |
-
"epoch": 0.
|
| 337 |
-
"grad_norm":
|
| 338 |
"learning_rate": 8.555555555555556e-06,
|
| 339 |
-
"loss": 0.
|
| 340 |
"step": 1150
|
| 341 |
},
|
| 342 |
{
|
| 343 |
-
"epoch": 0.
|
| 344 |
-
"grad_norm":
|
| 345 |
"learning_rate": 8.5e-06,
|
| 346 |
-
"loss": 0.
|
| 347 |
"step": 1175
|
| 348 |
},
|
| 349 |
{
|
| 350 |
-
"epoch": 0.
|
| 351 |
-
"grad_norm": 3.
|
| 352 |
"learning_rate": 8.444444444444446e-06,
|
| 353 |
-
"loss": 0.
|
| 354 |
"step": 1200
|
| 355 |
},
|
| 356 |
{
|
| 357 |
-
"epoch": 0.
|
| 358 |
-
"grad_norm": 3.
|
| 359 |
"learning_rate": 8.38888888888889e-06,
|
| 360 |
-
"loss": 0.
|
| 361 |
"step": 1225
|
| 362 |
},
|
| 363 |
{
|
| 364 |
-
"epoch": 0.
|
| 365 |
-
"grad_norm": 3.
|
| 366 |
"learning_rate": 8.333333333333334e-06,
|
| 367 |
-
"loss": 0.
|
| 368 |
"step": 1250
|
| 369 |
},
|
| 370 |
{
|
| 371 |
-
"epoch": 0.
|
| 372 |
-
"grad_norm":
|
| 373 |
"learning_rate": 8.277777777777778e-06,
|
| 374 |
-
"loss": 0.
|
| 375 |
"step": 1275
|
| 376 |
},
|
| 377 |
{
|
| 378 |
-
"epoch": 0.
|
| 379 |
-
"grad_norm": 3.
|
| 380 |
"learning_rate": 8.222222222222222e-06,
|
| 381 |
-
"loss": 0.
|
| 382 |
"step": 1300
|
| 383 |
},
|
| 384 |
{
|
| 385 |
-
"epoch": 0.
|
| 386 |
-
"grad_norm": 3.
|
| 387 |
"learning_rate": 8.166666666666668e-06,
|
| 388 |
-
"loss": 0.
|
| 389 |
"step": 1325
|
| 390 |
},
|
| 391 |
{
|
| 392 |
-
"epoch": 0.
|
| 393 |
-
"grad_norm": 3.
|
| 394 |
"learning_rate": 8.111111111111112e-06,
|
| 395 |
-
"loss": 0.
|
| 396 |
"step": 1350
|
| 397 |
},
|
| 398 |
{
|
| 399 |
-
"epoch": 0.
|
| 400 |
-
"grad_norm": 3.
|
| 401 |
"learning_rate": 8.055555555555557e-06,
|
| 402 |
-
"loss": 0.
|
| 403 |
"step": 1375
|
| 404 |
},
|
| 405 |
{
|
| 406 |
-
"epoch": 0.
|
| 407 |
-
"grad_norm":
|
| 408 |
"learning_rate": 8.000000000000001e-06,
|
| 409 |
-
"loss": 0.
|
| 410 |
"step": 1400
|
| 411 |
},
|
| 412 |
{
|
| 413 |
-
"epoch": 0.
|
| 414 |
-
"grad_norm":
|
| 415 |
"learning_rate": 7.944444444444445e-06,
|
| 416 |
-
"loss": 0.
|
| 417 |
"step": 1425
|
| 418 |
},
|
| 419 |
{
|
| 420 |
-
"epoch": 0.
|
| 421 |
-
"grad_norm": 3.
|
| 422 |
"learning_rate": 7.88888888888889e-06,
|
| 423 |
-
"loss": 0.
|
| 424 |
"step": 1450
|
| 425 |
},
|
| 426 |
{
|
| 427 |
-
"epoch": 0.
|
| 428 |
-
"grad_norm":
|
| 429 |
"learning_rate": 7.833333333333333e-06,
|
| 430 |
-
"loss": 0.
|
| 431 |
"step": 1475
|
| 432 |
},
|
| 433 |
{
|
| 434 |
-
"epoch": 0.
|
| 435 |
-
"grad_norm": 3.
|
| 436 |
"learning_rate": 7.77777777777778e-06,
|
| 437 |
-
"loss": 0.
|
| 438 |
"step": 1500
|
| 439 |
},
|
| 440 |
{
|
| 441 |
-
"epoch": 0.
|
| 442 |
-
"grad_norm":
|
| 443 |
"learning_rate": 7.722222222222223e-06,
|
| 444 |
-
"loss": 0.
|
| 445 |
"step": 1525
|
| 446 |
},
|
| 447 |
{
|
| 448 |
-
"epoch": 0.
|
| 449 |
-
"grad_norm": 3.
|
| 450 |
"learning_rate": 7.666666666666667e-06,
|
| 451 |
-
"loss": 0.
|
| 452 |
"step": 1550
|
| 453 |
},
|
| 454 |
{
|
| 455 |
-
"epoch": 0.
|
| 456 |
-
"grad_norm": 3.
|
| 457 |
"learning_rate": 7.611111111111111e-06,
|
| 458 |
-
"loss": 0.
|
| 459 |
"step": 1575
|
| 460 |
},
|
| 461 |
{
|
| 462 |
-
"epoch": 0.
|
| 463 |
-
"grad_norm": 3.
|
| 464 |
"learning_rate": 7.555555555555556e-06,
|
| 465 |
-
"loss": 0.
|
| 466 |
"step": 1600
|
| 467 |
},
|
| 468 |
{
|
| 469 |
-
"epoch": 0.
|
| 470 |
-
"grad_norm": 3.
|
| 471 |
"learning_rate": 7.500000000000001e-06,
|
| 472 |
-
"loss": 0.
|
| 473 |
"step": 1625
|
| 474 |
},
|
| 475 |
{
|
| 476 |
-
"epoch": 0.
|
| 477 |
-
"grad_norm": 3.
|
| 478 |
"learning_rate": 7.444444444444445e-06,
|
| 479 |
-
"loss": 0.
|
| 480 |
"step": 1650
|
| 481 |
},
|
| 482 |
{
|
| 483 |
-
"epoch": 0.
|
| 484 |
-
"grad_norm": 3.
|
| 485 |
"learning_rate": 7.38888888888889e-06,
|
| 486 |
-
"loss": 0.
|
| 487 |
"step": 1675
|
| 488 |
},
|
| 489 |
{
|
| 490 |
-
"epoch": 0.
|
| 491 |
-
"grad_norm": 3.
|
| 492 |
"learning_rate": 7.333333333333333e-06,
|
| 493 |
-
"loss": 0.
|
| 494 |
"step": 1700
|
| 495 |
},
|
| 496 |
{
|
| 497 |
-
"epoch": 0.
|
| 498 |
-
"grad_norm":
|
| 499 |
"learning_rate": 7.277777777777778e-06,
|
| 500 |
-
"loss": 0.
|
| 501 |
"step": 1725
|
| 502 |
},
|
| 503 |
{
|
| 504 |
-
"epoch": 0.
|
| 505 |
-
"grad_norm": 3.
|
| 506 |
"learning_rate": 7.222222222222223e-06,
|
| 507 |
-
"loss": 0.
|
| 508 |
"step": 1750
|
| 509 |
},
|
| 510 |
{
|
| 511 |
-
"epoch":
|
| 512 |
-
"grad_norm":
|
| 513 |
"learning_rate": 7.166666666666667e-06,
|
| 514 |
-
"loss": 0.
|
| 515 |
"step": 1775
|
| 516 |
},
|
| 517 |
{
|
| 518 |
-
"epoch":
|
| 519 |
-
"grad_norm": 2.
|
| 520 |
"learning_rate": 7.111111111111112e-06,
|
| 521 |
-
"loss": 0.
|
| 522 |
"step": 1800
|
| 523 |
},
|
| 524 |
{
|
| 525 |
-
"epoch":
|
| 526 |
-
"grad_norm":
|
| 527 |
"learning_rate": 7.055555555555557e-06,
|
| 528 |
-
"loss": 0.
|
| 529 |
"step": 1825
|
| 530 |
},
|
| 531 |
{
|
| 532 |
-
"epoch":
|
| 533 |
-
"grad_norm":
|
| 534 |
"learning_rate": 7e-06,
|
| 535 |
-
"loss": 0.
|
| 536 |
"step": 1850
|
| 537 |
},
|
| 538 |
{
|
| 539 |
-
"epoch":
|
| 540 |
-
"grad_norm":
|
| 541 |
"learning_rate": 6.944444444444445e-06,
|
| 542 |
-
"loss": 0.
|
| 543 |
"step": 1875
|
| 544 |
},
|
| 545 |
{
|
| 546 |
-
"epoch":
|
| 547 |
-
"grad_norm":
|
| 548 |
"learning_rate": 6.88888888888889e-06,
|
| 549 |
-
"loss": 0.
|
| 550 |
"step": 1900
|
| 551 |
},
|
| 552 |
{
|
| 553 |
-
"epoch":
|
| 554 |
-
"grad_norm":
|
| 555 |
"learning_rate": 6.833333333333334e-06,
|
| 556 |
-
"loss": 0.
|
| 557 |
"step": 1925
|
| 558 |
},
|
| 559 |
{
|
| 560 |
-
"epoch":
|
| 561 |
-
"grad_norm": 3.
|
| 562 |
"learning_rate": 6.777777777777779e-06,
|
| 563 |
-
"loss": 0.
|
| 564 |
"step": 1950
|
| 565 |
},
|
| 566 |
{
|
| 567 |
-
"epoch":
|
| 568 |
-
"grad_norm":
|
| 569 |
"learning_rate": 6.7222222222222235e-06,
|
| 570 |
-
"loss": 0.
|
| 571 |
"step": 1975
|
| 572 |
},
|
| 573 |
{
|
| 574 |
-
"epoch":
|
| 575 |
-
"grad_norm":
|
| 576 |
"learning_rate": 6.666666666666667e-06,
|
| 577 |
-
"loss": 0.
|
| 578 |
"step": 2000
|
| 579 |
},
|
| 580 |
{
|
| 581 |
-
"epoch":
|
| 582 |
-
"eval_loss": 0.
|
| 583 |
-
"eval_runtime":
|
| 584 |
-
"eval_samples_per_second":
|
| 585 |
-
"eval_steps_per_second": 0.
|
| 586 |
-
"eval_wer": 0.
|
| 587 |
"step": 2000
|
| 588 |
},
|
| 589 |
{
|
| 590 |
-
"epoch":
|
| 591 |
-
"grad_norm":
|
| 592 |
"learning_rate": 6.6111111111111115e-06,
|
| 593 |
-
"loss": 0.
|
| 594 |
"step": 2025
|
| 595 |
},
|
| 596 |
{
|
| 597 |
-
"epoch":
|
| 598 |
-
"grad_norm":
|
| 599 |
"learning_rate": 6.555555555555556e-06,
|
| 600 |
-
"loss": 0.
|
| 601 |
"step": 2050
|
| 602 |
},
|
| 603 |
{
|
| 604 |
-
"epoch":
|
| 605 |
-
"grad_norm":
|
| 606 |
"learning_rate": 6.5000000000000004e-06,
|
| 607 |
-
"loss": 0.
|
| 608 |
"step": 2075
|
| 609 |
},
|
| 610 |
{
|
| 611 |
-
"epoch":
|
| 612 |
-
"grad_norm":
|
| 613 |
"learning_rate": 6.444444444444445e-06,
|
| 614 |
-
"loss": 0.
|
| 615 |
"step": 2100
|
| 616 |
},
|
| 617 |
{
|
| 618 |
-
"epoch":
|
| 619 |
-
"grad_norm":
|
| 620 |
"learning_rate": 6.3888888888888885e-06,
|
| 621 |
-
"loss": 0.
|
| 622 |
"step": 2125
|
| 623 |
},
|
| 624 |
{
|
| 625 |
-
"epoch":
|
| 626 |
-
"grad_norm": 3.
|
| 627 |
"learning_rate": 6.333333333333333e-06,
|
| 628 |
-
"loss": 0.
|
| 629 |
"step": 2150
|
| 630 |
},
|
| 631 |
{
|
| 632 |
-
"epoch":
|
| 633 |
-
"grad_norm": 3.
|
| 634 |
"learning_rate": 6.277777777777778e-06,
|
| 635 |
-
"loss": 0.
|
| 636 |
"step": 2175
|
| 637 |
},
|
| 638 |
{
|
| 639 |
-
"epoch":
|
| 640 |
-
"grad_norm":
|
| 641 |
"learning_rate": 6.222222222222223e-06,
|
| 642 |
-
"loss": 0.
|
| 643 |
"step": 2200
|
| 644 |
},
|
| 645 |
{
|
| 646 |
-
"epoch":
|
| 647 |
-
"grad_norm":
|
| 648 |
"learning_rate": 6.166666666666667e-06,
|
| 649 |
-
"loss": 0.
|
| 650 |
"step": 2225
|
| 651 |
},
|
| 652 |
{
|
| 653 |
-
"epoch":
|
| 654 |
-
"grad_norm":
|
| 655 |
"learning_rate": 6.111111111111112e-06,
|
| 656 |
-
"loss": 0.
|
| 657 |
"step": 2250
|
| 658 |
},
|
| 659 |
{
|
| 660 |
-
"epoch":
|
| 661 |
-
"grad_norm": 3.
|
| 662 |
"learning_rate": 6.055555555555555e-06,
|
| 663 |
-
"loss": 0.
|
| 664 |
"step": 2275
|
| 665 |
},
|
| 666 |
{
|
| 667 |
-
"epoch":
|
| 668 |
-
"grad_norm":
|
| 669 |
"learning_rate": 6e-06,
|
| 670 |
-
"loss": 0.
|
| 671 |
"step": 2300
|
| 672 |
},
|
| 673 |
{
|
| 674 |
-
"epoch":
|
| 675 |
-
"grad_norm":
|
| 676 |
"learning_rate": 5.944444444444445e-06,
|
| 677 |
-
"loss": 0.
|
| 678 |
"step": 2325
|
| 679 |
},
|
| 680 |
{
|
| 681 |
-
"epoch":
|
| 682 |
-
"grad_norm":
|
| 683 |
"learning_rate": 5.88888888888889e-06,
|
| 684 |
-
"loss": 0.
|
| 685 |
"step": 2350
|
| 686 |
},
|
| 687 |
{
|
| 688 |
-
"epoch":
|
| 689 |
-
"grad_norm":
|
| 690 |
"learning_rate": 5.833333333333334e-06,
|
| 691 |
-
"loss": 0.
|
| 692 |
"step": 2375
|
| 693 |
},
|
| 694 |
{
|
| 695 |
-
"epoch":
|
| 696 |
-
"grad_norm":
|
| 697 |
-
"learning_rate": 5.
|
| 698 |
-
"loss": 0.
|
| 699 |
"step": 2400
|
| 700 |
},
|
| 701 |
{
|
| 702 |
-
"epoch":
|
| 703 |
-
"grad_norm":
|
| 704 |
-
"learning_rate": 5.
|
| 705 |
-
"loss": 0.
|
| 706 |
"step": 2425
|
| 707 |
},
|
| 708 |
{
|
| 709 |
-
"epoch":
|
| 710 |
-
"grad_norm":
|
| 711 |
-
"learning_rate": 5.
|
| 712 |
-
"loss": 0.
|
| 713 |
"step": 2450
|
| 714 |
},
|
| 715 |
{
|
| 716 |
-
"epoch":
|
| 717 |
-
"grad_norm": 2.
|
| 718 |
-
"learning_rate": 5.
|
| 719 |
-
"loss": 0.
|
| 720 |
"step": 2475
|
| 721 |
},
|
| 722 |
{
|
| 723 |
-
"epoch":
|
| 724 |
-
"grad_norm":
|
| 725 |
-
"learning_rate": 5.
|
| 726 |
-
"loss": 0.
|
| 727 |
"step": 2500
|
| 728 |
},
|
| 729 |
{
|
| 730 |
-
"epoch":
|
| 731 |
-
"grad_norm":
|
| 732 |
-
"learning_rate": 5.
|
| 733 |
-
"loss": 0.
|
| 734 |
"step": 2525
|
| 735 |
},
|
| 736 |
{
|
| 737 |
-
"epoch":
|
| 738 |
-
"grad_norm":
|
| 739 |
-
"learning_rate": 5.
|
| 740 |
-
"loss": 0.
|
| 741 |
"step": 2550
|
| 742 |
},
|
| 743 |
{
|
| 744 |
-
"epoch":
|
| 745 |
-
"grad_norm": 3.
|
| 746 |
-
"learning_rate": 5.
|
| 747 |
-
"loss": 0.
|
| 748 |
"step": 2575
|
| 749 |
},
|
| 750 |
{
|
| 751 |
-
"epoch":
|
| 752 |
-
"grad_norm":
|
| 753 |
-
"learning_rate": 5.
|
| 754 |
-
"loss": 0.
|
| 755 |
"step": 2600
|
| 756 |
},
|
| 757 |
{
|
| 758 |
-
"epoch":
|
| 759 |
-
"grad_norm":
|
| 760 |
-
"learning_rate": 5.
|
| 761 |
-
"loss": 0.
|
| 762 |
"step": 2625
|
| 763 |
},
|
| 764 |
{
|
| 765 |
-
"epoch":
|
| 766 |
-
"grad_norm": 2.
|
| 767 |
-
"learning_rate": 5.
|
| 768 |
-
"loss": 0.
|
| 769 |
"step": 2650
|
| 770 |
},
|
| 771 |
{
|
| 772 |
-
"epoch":
|
| 773 |
-
"grad_norm":
|
| 774 |
-
"learning_rate": 5.
|
| 775 |
-
"loss": 0.
|
| 776 |
"step": 2675
|
| 777 |
},
|
| 778 |
{
|
| 779 |
-
"epoch":
|
| 780 |
-
"grad_norm":
|
| 781 |
-
"learning_rate": 5.
|
| 782 |
-
"loss": 0.
|
| 783 |
"step": 2700
|
| 784 |
},
|
| 785 |
{
|
| 786 |
-
"epoch":
|
| 787 |
-
"grad_norm": 2.
|
| 788 |
-
"learning_rate": 5.
|
| 789 |
-
"loss": 0.
|
| 790 |
"step": 2725
|
| 791 |
},
|
| 792 |
{
|
| 793 |
-
"epoch":
|
| 794 |
-
"grad_norm":
|
| 795 |
-
"learning_rate":
|
| 796 |
-
"loss": 0.
|
| 797 |
"step": 2750
|
| 798 |
},
|
| 799 |
{
|
| 800 |
-
"epoch":
|
| 801 |
-
"grad_norm":
|
| 802 |
-
"learning_rate": 4.
|
| 803 |
-
"loss": 0.
|
| 804 |
"step": 2775
|
| 805 |
},
|
| 806 |
{
|
| 807 |
-
"epoch":
|
| 808 |
-
"grad_norm":
|
| 809 |
-
"learning_rate": 4.
|
| 810 |
-
"loss": 0.
|
| 811 |
"step": 2800
|
| 812 |
},
|
| 813 |
{
|
| 814 |
-
"epoch":
|
| 815 |
-
"grad_norm":
|
| 816 |
-
"learning_rate": 4.
|
| 817 |
-
"loss": 0.
|
| 818 |
"step": 2825
|
| 819 |
},
|
| 820 |
{
|
| 821 |
-
"epoch":
|
| 822 |
-
"grad_norm":
|
| 823 |
-
"learning_rate": 4.
|
| 824 |
-
"loss": 0.
|
| 825 |
"step": 2850
|
| 826 |
},
|
| 827 |
{
|
| 828 |
-
"epoch":
|
| 829 |
-
"grad_norm":
|
| 830 |
-
"learning_rate": 4.
|
| 831 |
-
"loss": 0.
|
| 832 |
"step": 2875
|
| 833 |
},
|
| 834 |
{
|
| 835 |
-
"epoch":
|
| 836 |
-
"grad_norm":
|
| 837 |
-
"learning_rate": 4.
|
| 838 |
-
"loss": 0.
|
| 839 |
"step": 2900
|
| 840 |
},
|
| 841 |
{
|
| 842 |
-
"epoch":
|
| 843 |
-
"grad_norm":
|
| 844 |
-
"learning_rate": 4.
|
| 845 |
-
"loss": 0.
|
| 846 |
"step": 2925
|
| 847 |
},
|
| 848 |
{
|
| 849 |
-
"epoch":
|
| 850 |
-
"grad_norm":
|
| 851 |
-
"learning_rate": 4.
|
| 852 |
-
"loss": 0.
|
| 853 |
"step": 2950
|
| 854 |
},
|
| 855 |
{
|
| 856 |
-
"epoch":
|
| 857 |
-
"grad_norm":
|
| 858 |
-
"learning_rate": 4.
|
| 859 |
-
"loss": 0.
|
| 860 |
"step": 2975
|
| 861 |
},
|
| 862 |
{
|
| 863 |
-
"epoch":
|
| 864 |
-
"grad_norm": 2.
|
| 865 |
-
"learning_rate": 4.
|
| 866 |
-
"loss": 0.
|
| 867 |
"step": 3000
|
| 868 |
},
|
| 869 |
{
|
| 870 |
-
"epoch":
|
| 871 |
-
"eval_loss": 0.
|
| 872 |
-
"eval_runtime":
|
| 873 |
-
"eval_samples_per_second": 2.
|
| 874 |
-
"eval_steps_per_second": 0.
|
| 875 |
-
"eval_wer": 0.
|
| 876 |
"step": 3000
|
| 877 |
},
|
| 878 |
{
|
| 879 |
-
"epoch":
|
| 880 |
-
"grad_norm":
|
| 881 |
-
"learning_rate": 4.
|
| 882 |
-
"loss": 0.
|
| 883 |
"step": 3025
|
| 884 |
},
|
| 885 |
{
|
| 886 |
-
"epoch":
|
| 887 |
-
"grad_norm":
|
| 888 |
-
"learning_rate": 4.
|
| 889 |
-
"loss": 0.
|
| 890 |
"step": 3050
|
| 891 |
},
|
| 892 |
{
|
| 893 |
-
"epoch":
|
| 894 |
-
"grad_norm":
|
| 895 |
-
"learning_rate": 4.
|
| 896 |
-
"loss": 0.
|
| 897 |
"step": 3075
|
| 898 |
},
|
| 899 |
{
|
| 900 |
-
"epoch": 1.
|
| 901 |
-
"grad_norm":
|
| 902 |
-
"learning_rate": 4.
|
| 903 |
-
"loss": 0.
|
| 904 |
"step": 3100
|
| 905 |
},
|
| 906 |
{
|
| 907 |
-
"epoch": 1.
|
| 908 |
-
"grad_norm":
|
| 909 |
-
"learning_rate": 4.
|
| 910 |
-
"loss": 0.
|
| 911 |
"step": 3125
|
| 912 |
},
|
| 913 |
{
|
| 914 |
-
"epoch": 1.
|
| 915 |
-
"grad_norm":
|
| 916 |
-
"learning_rate": 4.
|
| 917 |
-
"loss": 0.
|
| 918 |
"step": 3150
|
| 919 |
},
|
| 920 |
{
|
| 921 |
-
"epoch": 1.
|
| 922 |
-
"grad_norm":
|
| 923 |
-
"learning_rate": 4.
|
| 924 |
-
"loss": 0.
|
| 925 |
"step": 3175
|
| 926 |
},
|
| 927 |
{
|
| 928 |
-
"epoch": 1.
|
| 929 |
-
"grad_norm":
|
| 930 |
-
"learning_rate": 4.
|
| 931 |
-
"loss": 0.
|
| 932 |
"step": 3200
|
| 933 |
},
|
| 934 |
{
|
| 935 |
-
"epoch": 1.
|
| 936 |
-
"grad_norm":
|
| 937 |
-
"learning_rate": 3.
|
| 938 |
-
"loss": 0.
|
| 939 |
"step": 3225
|
| 940 |
},
|
| 941 |
{
|
| 942 |
-
"epoch": 1.
|
| 943 |
-
"grad_norm":
|
| 944 |
-
"learning_rate": 3.
|
| 945 |
-
"loss": 0.
|
| 946 |
"step": 3250
|
| 947 |
},
|
| 948 |
{
|
| 949 |
-
"epoch": 1.
|
| 950 |
-
"grad_norm":
|
| 951 |
-
"learning_rate": 3.
|
| 952 |
-
"loss": 0.
|
| 953 |
"step": 3275
|
| 954 |
},
|
| 955 |
{
|
| 956 |
-
"epoch": 1.
|
| 957 |
-
"grad_norm":
|
| 958 |
-
"learning_rate": 3.
|
| 959 |
-
"loss": 0.
|
| 960 |
"step": 3300
|
| 961 |
},
|
| 962 |
{
|
| 963 |
-
"epoch": 1.
|
| 964 |
-
"grad_norm": 2.
|
| 965 |
-
"learning_rate": 3.
|
| 966 |
-
"loss": 0.
|
| 967 |
"step": 3325
|
| 968 |
},
|
| 969 |
{
|
| 970 |
-
"epoch": 1.
|
| 971 |
-
"grad_norm": 2.
|
| 972 |
-
"learning_rate": 3.
|
| 973 |
-
"loss": 0.
|
| 974 |
"step": 3350
|
| 975 |
},
|
| 976 |
{
|
| 977 |
-
"epoch": 1.
|
| 978 |
-
"grad_norm": 2.
|
| 979 |
-
"learning_rate": 3.
|
| 980 |
-
"loss": 0.
|
| 981 |
"step": 3375
|
| 982 |
},
|
| 983 |
{
|
| 984 |
-
"epoch": 1.
|
| 985 |
-
"grad_norm":
|
| 986 |
-
"learning_rate": 3.
|
| 987 |
-
"loss": 0.
|
| 988 |
"step": 3400
|
| 989 |
},
|
| 990 |
{
|
| 991 |
-
"epoch": 1.
|
| 992 |
-
"grad_norm": 2.
|
| 993 |
-
"learning_rate": 3.
|
| 994 |
-
"loss": 0.
|
| 995 |
"step": 3425
|
| 996 |
},
|
| 997 |
{
|
| 998 |
-
"epoch": 1.
|
| 999 |
-
"grad_norm":
|
| 1000 |
-
"learning_rate": 3.
|
| 1001 |
-
"loss": 0.
|
| 1002 |
"step": 3450
|
| 1003 |
},
|
| 1004 |
{
|
| 1005 |
-
"epoch": 1.
|
| 1006 |
-
"grad_norm":
|
| 1007 |
-
"learning_rate": 3.
|
| 1008 |
-
"loss": 0.
|
| 1009 |
"step": 3475
|
| 1010 |
},
|
| 1011 |
{
|
| 1012 |
-
"epoch": 1.
|
| 1013 |
-
"grad_norm": 2.
|
| 1014 |
-
"learning_rate": 3.
|
| 1015 |
-
"loss": 0.
|
| 1016 |
"step": 3500
|
| 1017 |
},
|
| 1018 |
{
|
| 1019 |
-
"epoch":
|
| 1020 |
-
"grad_norm": 2.
|
| 1021 |
-
"learning_rate": 3.
|
| 1022 |
-
"loss": 0.
|
| 1023 |
"step": 3525
|
| 1024 |
},
|
| 1025 |
{
|
| 1026 |
-
"epoch":
|
| 1027 |
-
"grad_norm": 2.
|
| 1028 |
-
"learning_rate": 3.
|
| 1029 |
-
"loss": 0.
|
| 1030 |
"step": 3550
|
| 1031 |
},
|
| 1032 |
{
|
| 1033 |
-
"epoch":
|
| 1034 |
-
"grad_norm": 2.
|
| 1035 |
-
"learning_rate": 3.
|
| 1036 |
-
"loss": 0.
|
| 1037 |
"step": 3575
|
| 1038 |
},
|
| 1039 |
{
|
| 1040 |
-
"epoch":
|
| 1041 |
-
"grad_norm":
|
| 1042 |
-
"learning_rate": 3.
|
| 1043 |
-
"loss": 0.
|
| 1044 |
"step": 3600
|
| 1045 |
},
|
| 1046 |
{
|
| 1047 |
-
"epoch":
|
| 1048 |
-
"grad_norm":
|
| 1049 |
-
"learning_rate": 3.
|
| 1050 |
-
"loss": 0.
|
| 1051 |
"step": 3625
|
| 1052 |
},
|
| 1053 |
{
|
| 1054 |
-
"epoch":
|
| 1055 |
-
"grad_norm": 2.
|
| 1056 |
-
"learning_rate":
|
| 1057 |
-
"loss": 0.
|
| 1058 |
"step": 3650
|
| 1059 |
},
|
| 1060 |
{
|
| 1061 |
-
"epoch":
|
| 1062 |
-
"grad_norm":
|
| 1063 |
-
"learning_rate": 2.
|
| 1064 |
-
"loss": 0.
|
| 1065 |
"step": 3675
|
| 1066 |
},
|
| 1067 |
{
|
| 1068 |
-
"epoch":
|
| 1069 |
-
"grad_norm": 2.
|
| 1070 |
-
"learning_rate": 2.
|
| 1071 |
-
"loss": 0.
|
| 1072 |
"step": 3700
|
| 1073 |
},
|
| 1074 |
{
|
| 1075 |
-
"epoch":
|
| 1076 |
-
"grad_norm": 2.
|
| 1077 |
-
"learning_rate": 2.
|
| 1078 |
-
"loss": 0.
|
| 1079 |
"step": 3725
|
| 1080 |
},
|
| 1081 |
{
|
| 1082 |
-
"epoch":
|
| 1083 |
-
"grad_norm": 2.
|
| 1084 |
-
"learning_rate": 2.
|
| 1085 |
-
"loss": 0.
|
| 1086 |
"step": 3750
|
| 1087 |
},
|
| 1088 |
{
|
| 1089 |
-
"epoch":
|
| 1090 |
-
"grad_norm":
|
| 1091 |
-
"learning_rate": 2.
|
| 1092 |
-
"loss": 0.
|
| 1093 |
"step": 3775
|
| 1094 |
},
|
| 1095 |
{
|
| 1096 |
-
"epoch":
|
| 1097 |
-
"grad_norm": 2.
|
| 1098 |
-
"learning_rate": 2.
|
| 1099 |
-
"loss": 0.
|
| 1100 |
"step": 3800
|
| 1101 |
},
|
| 1102 |
{
|
| 1103 |
-
"epoch":
|
| 1104 |
-
"grad_norm":
|
| 1105 |
-
"learning_rate": 2.
|
| 1106 |
-
"loss": 0.
|
| 1107 |
"step": 3825
|
| 1108 |
},
|
| 1109 |
{
|
| 1110 |
-
"epoch":
|
| 1111 |
-
"grad_norm":
|
| 1112 |
-
"learning_rate": 2.
|
| 1113 |
-
"loss": 0.
|
| 1114 |
"step": 3850
|
| 1115 |
},
|
| 1116 |
{
|
| 1117 |
-
"epoch":
|
| 1118 |
-
"grad_norm": 2.
|
| 1119 |
-
"learning_rate": 2.
|
| 1120 |
-
"loss": 0.
|
| 1121 |
"step": 3875
|
| 1122 |
},
|
| 1123 |
{
|
| 1124 |
-
"epoch":
|
| 1125 |
-
"grad_norm":
|
| 1126 |
-
"learning_rate": 2.
|
| 1127 |
-
"loss": 0.
|
| 1128 |
"step": 3900
|
| 1129 |
},
|
| 1130 |
{
|
| 1131 |
-
"epoch":
|
| 1132 |
-
"grad_norm":
|
| 1133 |
-
"learning_rate": 2.
|
| 1134 |
-
"loss": 0.
|
| 1135 |
"step": 3925
|
| 1136 |
},
|
| 1137 |
{
|
| 1138 |
-
"epoch":
|
| 1139 |
-
"grad_norm":
|
| 1140 |
-
"learning_rate": 2.
|
| 1141 |
-
"loss": 0.
|
| 1142 |
"step": 3950
|
| 1143 |
},
|
| 1144 |
{
|
| 1145 |
-
"epoch":
|
| 1146 |
-
"grad_norm": 2.
|
| 1147 |
-
"learning_rate": 2.
|
| 1148 |
-
"loss": 0.
|
| 1149 |
"step": 3975
|
| 1150 |
},
|
| 1151 |
{
|
| 1152 |
-
"epoch":
|
| 1153 |
-
"grad_norm":
|
| 1154 |
-
"learning_rate": 2.
|
| 1155 |
-
"loss": 0.
|
| 1156 |
"step": 4000
|
| 1157 |
},
|
| 1158 |
{
|
| 1159 |
-
"epoch":
|
| 1160 |
-
"eval_loss": 0.
|
| 1161 |
-
"eval_runtime":
|
| 1162 |
-
"eval_samples_per_second": 2.
|
| 1163 |
-
"eval_steps_per_second": 0.
|
| 1164 |
-
"eval_wer": 0.
|
| 1165 |
"step": 4000
|
| 1166 |
},
|
| 1167 |
{
|
| 1168 |
-
"epoch":
|
| 1169 |
-
"grad_norm": 2.
|
| 1170 |
-
"learning_rate": 2.
|
| 1171 |
-
"loss": 0.
|
| 1172 |
"step": 4025
|
| 1173 |
},
|
| 1174 |
{
|
| 1175 |
-
"epoch":
|
| 1176 |
-
"grad_norm": 2.
|
| 1177 |
-
"learning_rate": 2.
|
| 1178 |
-
"loss": 0.
|
| 1179 |
"step": 4050
|
| 1180 |
},
|
| 1181 |
{
|
| 1182 |
-
"epoch":
|
| 1183 |
-
"grad_norm": 2.
|
| 1184 |
-
"learning_rate": 2.
|
| 1185 |
-
"loss": 0.
|
| 1186 |
"step": 4075
|
| 1187 |
},
|
| 1188 |
{
|
| 1189 |
-
"epoch":
|
| 1190 |
-
"grad_norm":
|
| 1191 |
-
"learning_rate": 2.
|
| 1192 |
-
"loss": 0.
|
| 1193 |
"step": 4100
|
| 1194 |
},
|
| 1195 |
{
|
| 1196 |
-
"epoch":
|
| 1197 |
-
"grad_norm": 2.
|
| 1198 |
-
"learning_rate": 1.
|
| 1199 |
-
"loss": 0.
|
| 1200 |
"step": 4125
|
| 1201 |
},
|
| 1202 |
{
|
| 1203 |
-
"epoch":
|
| 1204 |
-
"grad_norm": 2.
|
| 1205 |
-
"learning_rate": 1.
|
| 1206 |
-
"loss": 0.
|
| 1207 |
"step": 4150
|
| 1208 |
},
|
| 1209 |
{
|
| 1210 |
-
"epoch":
|
| 1211 |
-
"grad_norm": 2.
|
| 1212 |
-
"learning_rate": 1.
|
| 1213 |
-
"loss": 0.
|
| 1214 |
"step": 4175
|
| 1215 |
},
|
| 1216 |
{
|
| 1217 |
-
"epoch":
|
| 1218 |
-
"grad_norm": 2.
|
| 1219 |
-
"learning_rate": 1.
|
| 1220 |
-
"loss": 0.
|
| 1221 |
"step": 4200
|
| 1222 |
},
|
| 1223 |
{
|
| 1224 |
-
"epoch":
|
| 1225 |
-
"grad_norm": 2.
|
| 1226 |
-
"learning_rate": 1.
|
| 1227 |
-
"loss": 0.
|
| 1228 |
"step": 4225
|
| 1229 |
},
|
| 1230 |
{
|
| 1231 |
-
"epoch":
|
| 1232 |
-
"grad_norm": 2.
|
| 1233 |
-
"learning_rate": 1.
|
| 1234 |
-
"loss": 0.
|
| 1235 |
"step": 4250
|
| 1236 |
},
|
| 1237 |
{
|
| 1238 |
-
"epoch":
|
| 1239 |
-
"grad_norm": 2.
|
| 1240 |
-
"learning_rate": 1.
|
| 1241 |
-
"loss": 0.
|
| 1242 |
"step": 4275
|
| 1243 |
},
|
| 1244 |
{
|
| 1245 |
-
"epoch":
|
| 1246 |
-
"grad_norm":
|
| 1247 |
-
"learning_rate": 1.
|
| 1248 |
-
"loss": 0.
|
| 1249 |
"step": 4300
|
| 1250 |
},
|
| 1251 |
{
|
| 1252 |
-
"epoch":
|
| 1253 |
-
"grad_norm": 2.
|
| 1254 |
-
"learning_rate": 1.
|
| 1255 |
-
"loss": 0.
|
| 1256 |
"step": 4325
|
| 1257 |
},
|
| 1258 |
{
|
| 1259 |
-
"epoch":
|
| 1260 |
-
"grad_norm": 2.
|
| 1261 |
-
"learning_rate": 1.
|
| 1262 |
-
"loss": 0.
|
| 1263 |
"step": 4350
|
| 1264 |
},
|
| 1265 |
{
|
| 1266 |
-
"epoch":
|
| 1267 |
-
"grad_norm": 2.
|
| 1268 |
-
"learning_rate": 1.
|
| 1269 |
-
"loss": 0.
|
| 1270 |
"step": 4375
|
| 1271 |
},
|
| 1272 |
{
|
| 1273 |
-
"epoch":
|
| 1274 |
-
"grad_norm": 2.
|
| 1275 |
-
"learning_rate": 1.
|
| 1276 |
-
"loss": 0.
|
| 1277 |
"step": 4400
|
| 1278 |
},
|
| 1279 |
{
|
| 1280 |
-
"epoch":
|
| 1281 |
-
"grad_norm": 2.
|
| 1282 |
"learning_rate": 1.28e-06,
|
| 1283 |
-
"loss": 0.
|
| 1284 |
"step": 4425
|
| 1285 |
},
|
| 1286 |
{
|
| 1287 |
-
"epoch":
|
| 1288 |
-
"grad_norm":
|
| 1289 |
"learning_rate": 1.2244444444444445e-06,
|
| 1290 |
-
"loss": 0.
|
| 1291 |
"step": 4450
|
| 1292 |
},
|
| 1293 |
{
|
| 1294 |
-
"epoch":
|
| 1295 |
-
"grad_norm":
|
| 1296 |
"learning_rate": 1.168888888888889e-06,
|
| 1297 |
-
"loss": 0.
|
| 1298 |
"step": 4475
|
| 1299 |
},
|
| 1300 |
{
|
| 1301 |
-
"epoch":
|
| 1302 |
-
"grad_norm":
|
| 1303 |
-
"learning_rate": 1.
|
| 1304 |
-
"loss": 0.
|
| 1305 |
"step": 4500
|
| 1306 |
},
|
| 1307 |
{
|
| 1308 |
-
"epoch":
|
| 1309 |
-
"grad_norm": 2.
|
| 1310 |
-
"learning_rate": 1.
|
| 1311 |
-
"loss": 0.
|
| 1312 |
"step": 4525
|
| 1313 |
},
|
| 1314 |
{
|
| 1315 |
-
"epoch":
|
| 1316 |
-
"grad_norm":
|
| 1317 |
-
"learning_rate": 1.
|
| 1318 |
-
"loss": 0.
|
| 1319 |
"step": 4550
|
| 1320 |
},
|
| 1321 |
{
|
| 1322 |
-
"epoch":
|
| 1323 |
-
"grad_norm": 2.
|
| 1324 |
-
"learning_rate": 9.
|
| 1325 |
-
"loss": 0.
|
| 1326 |
"step": 4575
|
| 1327 |
},
|
| 1328 |
{
|
| 1329 |
-
"epoch":
|
| 1330 |
-
"grad_norm": 2.
|
| 1331 |
-
"learning_rate": 8.
|
| 1332 |
-
"loss": 0.
|
| 1333 |
"step": 4600
|
| 1334 |
},
|
| 1335 |
{
|
| 1336 |
-
"epoch":
|
| 1337 |
-
"grad_norm": 2.
|
| 1338 |
-
"learning_rate": 8.
|
| 1339 |
-
"loss": 0.
|
| 1340 |
"step": 4625
|
| 1341 |
},
|
| 1342 |
{
|
| 1343 |
-
"epoch":
|
| 1344 |
-
"grad_norm":
|
| 1345 |
-
"learning_rate": 7.
|
| 1346 |
-
"loss": 0.
|
| 1347 |
"step": 4650
|
| 1348 |
},
|
| 1349 |
{
|
| 1350 |
-
"epoch":
|
| 1351 |
-
"grad_norm": 2.
|
| 1352 |
-
"learning_rate": 7.
|
| 1353 |
-
"loss": 0.
|
| 1354 |
"step": 4675
|
| 1355 |
},
|
| 1356 |
{
|
| 1357 |
-
"epoch":
|
| 1358 |
-
"grad_norm": 2.
|
| 1359 |
-
"learning_rate": 6.
|
| 1360 |
-
"loss": 0.
|
| 1361 |
"step": 4700
|
| 1362 |
},
|
| 1363 |
{
|
| 1364 |
-
"epoch":
|
| 1365 |
-
"grad_norm": 2.
|
| 1366 |
-
"learning_rate": 6.
|
| 1367 |
-
"loss": 0.
|
| 1368 |
"step": 4725
|
| 1369 |
},
|
| 1370 |
{
|
| 1371 |
-
"epoch":
|
| 1372 |
-
"grad_norm":
|
| 1373 |
-
"learning_rate": 5.
|
| 1374 |
-
"loss": 0.
|
| 1375 |
"step": 4750
|
| 1376 |
},
|
| 1377 |
{
|
| 1378 |
-
"epoch":
|
| 1379 |
-
"grad_norm": 2.
|
| 1380 |
-
"learning_rate": 5.
|
| 1381 |
-
"loss": 0.
|
| 1382 |
"step": 4775
|
| 1383 |
},
|
| 1384 |
{
|
| 1385 |
-
"epoch":
|
| 1386 |
-
"grad_norm": 2.
|
| 1387 |
-
"learning_rate": 4.
|
| 1388 |
-
"loss": 0.
|
| 1389 |
"step": 4800
|
| 1390 |
},
|
| 1391 |
{
|
| 1392 |
-
"epoch":
|
| 1393 |
-
"grad_norm":
|
| 1394 |
-
"learning_rate": 3.
|
| 1395 |
-
"loss": 0.
|
| 1396 |
"step": 4825
|
| 1397 |
},
|
| 1398 |
{
|
| 1399 |
-
"epoch":
|
| 1400 |
-
"grad_norm": 2.
|
| 1401 |
-
"learning_rate": 3.
|
| 1402 |
-
"loss": 0.
|
| 1403 |
"step": 4850
|
| 1404 |
},
|
| 1405 |
{
|
| 1406 |
-
"epoch":
|
| 1407 |
-
"grad_norm":
|
| 1408 |
-
"learning_rate": 2.
|
| 1409 |
-
"loss": 0.
|
| 1410 |
"step": 4875
|
| 1411 |
},
|
| 1412 |
{
|
| 1413 |
-
"epoch":
|
| 1414 |
-
"grad_norm": 2.
|
| 1415 |
-
"learning_rate": 2.
|
| 1416 |
-
"loss": 0.
|
| 1417 |
"step": 4900
|
| 1418 |
},
|
| 1419 |
{
|
| 1420 |
-
"epoch":
|
| 1421 |
-
"grad_norm": 2.
|
| 1422 |
-
"learning_rate": 1.
|
| 1423 |
-
"loss": 0.
|
| 1424 |
"step": 4925
|
| 1425 |
},
|
| 1426 |
{
|
| 1427 |
-
"epoch":
|
| 1428 |
-
"grad_norm": 2.
|
| 1429 |
-
"learning_rate": 1.
|
| 1430 |
-
"loss": 0.
|
| 1431 |
"step": 4950
|
| 1432 |
},
|
| 1433 |
{
|
| 1434 |
-
"epoch":
|
| 1435 |
-
"grad_norm": 2.
|
| 1436 |
-
"learning_rate":
|
| 1437 |
-
"loss": 0.
|
| 1438 |
"step": 4975
|
| 1439 |
},
|
| 1440 |
{
|
| 1441 |
-
"epoch":
|
| 1442 |
-
"grad_norm": 2.
|
| 1443 |
-
"learning_rate":
|
| 1444 |
-
"loss": 0.
|
| 1445 |
"step": 5000
|
| 1446 |
},
|
| 1447 |
{
|
| 1448 |
-
"epoch":
|
| 1449 |
-
"eval_loss": 0.
|
| 1450 |
-
"eval_runtime":
|
| 1451 |
-
"eval_samples_per_second": 2.
|
| 1452 |
-
"eval_steps_per_second": 0.
|
| 1453 |
-
"eval_wer": 0.
|
| 1454 |
"step": 5000
|
| 1455 |
},
|
| 1456 |
{
|
| 1457 |
-
"epoch":
|
| 1458 |
"step": 5000,
|
| 1459 |
-
"total_flos": 5.
|
| 1460 |
-
"train_loss": 0.
|
| 1461 |
-
"train_runtime":
|
| 1462 |
-
"train_samples_per_second": 2.
|
| 1463 |
-
"train_steps_per_second": 0.
|
| 1464 |
}
|
| 1465 |
],
|
| 1466 |
"logging_steps": 25,
|
| 1467 |
"max_steps": 5000,
|
| 1468 |
"num_input_tokens_seen": 0,
|
| 1469 |
-
"num_train_epochs":
|
| 1470 |
"save_steps": 1000,
|
| 1471 |
"stateful_callbacks": {
|
| 1472 |
"TrainerControl": {
|
|
@@ -1480,7 +1480,7 @@
|
|
| 1480 |
"attributes": {}
|
| 1481 |
}
|
| 1482 |
},
|
| 1483 |
-
"total_flos": 5.
|
| 1484 |
"train_batch_size": 16,
|
| 1485 |
"trial_name": null,
|
| 1486 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.619795885306982,
|
| 5 |
"eval_steps": 1000,
|
| 6 |
"global_step": 5000,
|
| 7 |
"is_hyper_param_search": false,
|
|
|
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.008099789405475458,
|
| 13 |
+
"grad_norm": 8.217827796936035,
|
| 14 |
"learning_rate": 5.000000000000001e-07,
|
| 15 |
+
"loss": 1.6162,
|
| 16 |
"step": 25
|
| 17 |
},
|
| 18 |
{
|
| 19 |
+
"epoch": 0.016199578810950917,
|
| 20 |
+
"grad_norm": 5.221391677856445,
|
| 21 |
"learning_rate": 1.0000000000000002e-06,
|
| 22 |
+
"loss": 1.3426,
|
| 23 |
"step": 50
|
| 24 |
},
|
| 25 |
{
|
| 26 |
+
"epoch": 0.02429936821642637,
|
| 27 |
+
"grad_norm": 6.4570441246032715,
|
| 28 |
"learning_rate": 1.5e-06,
|
| 29 |
+
"loss": 1.0591,
|
| 30 |
"step": 75
|
| 31 |
},
|
| 32 |
{
|
| 33 |
+
"epoch": 0.03239915762190183,
|
| 34 |
+
"grad_norm": 4.746372699737549,
|
| 35 |
"learning_rate": 2.0000000000000003e-06,
|
| 36 |
+
"loss": 0.9419,
|
| 37 |
"step": 100
|
| 38 |
},
|
| 39 |
{
|
| 40 |
+
"epoch": 0.04049894702737729,
|
| 41 |
+
"grad_norm": 5.625121593475342,
|
| 42 |
"learning_rate": 2.5e-06,
|
| 43 |
+
"loss": 0.8389,
|
| 44 |
"step": 125
|
| 45 |
},
|
| 46 |
{
|
| 47 |
+
"epoch": 0.04859873643285274,
|
| 48 |
+
"grad_norm": 6.477604866027832,
|
| 49 |
"learning_rate": 3e-06,
|
| 50 |
+
"loss": 0.8019,
|
| 51 |
"step": 150
|
| 52 |
},
|
| 53 |
{
|
| 54 |
+
"epoch": 0.056698525838328205,
|
| 55 |
+
"grad_norm": 4.935534954071045,
|
| 56 |
"learning_rate": 3.5e-06,
|
| 57 |
+
"loss": 0.8007,
|
| 58 |
"step": 175
|
| 59 |
},
|
| 60 |
{
|
| 61 |
+
"epoch": 0.06479831524380367,
|
| 62 |
+
"grad_norm": 5.263591289520264,
|
| 63 |
"learning_rate": 4.000000000000001e-06,
|
| 64 |
+
"loss": 0.7419,
|
| 65 |
"step": 200
|
| 66 |
},
|
| 67 |
{
|
| 68 |
+
"epoch": 0.07289810464927912,
|
| 69 |
+
"grad_norm": 5.378949165344238,
|
| 70 |
"learning_rate": 4.5e-06,
|
| 71 |
+
"loss": 0.7391,
|
| 72 |
"step": 225
|
| 73 |
},
|
| 74 |
{
|
| 75 |
+
"epoch": 0.08099789405475458,
|
| 76 |
+
"grad_norm": 5.626790523529053,
|
| 77 |
"learning_rate": 5e-06,
|
| 78 |
+
"loss": 0.7243,
|
| 79 |
"step": 250
|
| 80 |
},
|
| 81 |
{
|
| 82 |
+
"epoch": 0.08909768346023003,
|
| 83 |
+
"grad_norm": 5.457970142364502,
|
| 84 |
"learning_rate": 5.500000000000001e-06,
|
| 85 |
+
"loss": 0.7273,
|
| 86 |
"step": 275
|
| 87 |
},
|
| 88 |
{
|
| 89 |
+
"epoch": 0.09719747286570549,
|
| 90 |
+
"grad_norm": 4.948482036590576,
|
| 91 |
"learning_rate": 6e-06,
|
| 92 |
+
"loss": 0.7105,
|
| 93 |
"step": 300
|
| 94 |
},
|
| 95 |
{
|
| 96 |
+
"epoch": 0.10529726227118096,
|
| 97 |
+
"grad_norm": 4.352142333984375,
|
| 98 |
"learning_rate": 6.5000000000000004e-06,
|
| 99 |
+
"loss": 0.7075,
|
| 100 |
"step": 325
|
| 101 |
},
|
| 102 |
{
|
| 103 |
+
"epoch": 0.11339705167665641,
|
| 104 |
+
"grad_norm": 4.84460973739624,
|
| 105 |
"learning_rate": 7e-06,
|
| 106 |
+
"loss": 0.6808,
|
| 107 |
"step": 350
|
| 108 |
},
|
| 109 |
{
|
| 110 |
+
"epoch": 0.12149684108213187,
|
| 111 |
+
"grad_norm": 5.333044052124023,
|
| 112 |
"learning_rate": 7.500000000000001e-06,
|
| 113 |
+
"loss": 0.6855,
|
| 114 |
"step": 375
|
| 115 |
},
|
| 116 |
{
|
| 117 |
+
"epoch": 0.12959663048760733,
|
| 118 |
+
"grad_norm": 5.835122585296631,
|
| 119 |
"learning_rate": 8.000000000000001e-06,
|
| 120 |
+
"loss": 0.642,
|
| 121 |
"step": 400
|
| 122 |
},
|
| 123 |
{
|
| 124 |
+
"epoch": 0.1376964198930828,
|
| 125 |
+
"grad_norm": 6.060296058654785,
|
| 126 |
"learning_rate": 8.5e-06,
|
| 127 |
+
"loss": 0.6443,
|
| 128 |
"step": 425
|
| 129 |
},
|
| 130 |
{
|
| 131 |
+
"epoch": 0.14579620929855824,
|
| 132 |
+
"grad_norm": 4.375275135040283,
|
| 133 |
"learning_rate": 9e-06,
|
| 134 |
+
"loss": 0.6477,
|
| 135 |
"step": 450
|
| 136 |
},
|
| 137 |
{
|
| 138 |
+
"epoch": 0.1538959987040337,
|
| 139 |
+
"grad_norm": 4.223667144775391,
|
| 140 |
"learning_rate": 9.5e-06,
|
| 141 |
+
"loss": 0.6369,
|
| 142 |
"step": 475
|
| 143 |
},
|
| 144 |
{
|
| 145 |
+
"epoch": 0.16199578810950915,
|
| 146 |
+
"grad_norm": 10.034436225891113,
|
| 147 |
"learning_rate": 1e-05,
|
| 148 |
+
"loss": 0.6853,
|
| 149 |
"step": 500
|
| 150 |
},
|
| 151 |
{
|
| 152 |
+
"epoch": 0.1700955775149846,
|
| 153 |
+
"grad_norm": 4.590435028076172,
|
| 154 |
"learning_rate": 9.944444444444445e-06,
|
| 155 |
+
"loss": 0.6149,
|
| 156 |
"step": 525
|
| 157 |
},
|
| 158 |
{
|
| 159 |
+
"epoch": 0.17819536692046006,
|
| 160 |
+
"grad_norm": 4.104828357696533,
|
| 161 |
"learning_rate": 9.88888888888889e-06,
|
| 162 |
+
"loss": 0.6267,
|
| 163 |
"step": 550
|
| 164 |
},
|
| 165 |
{
|
| 166 |
+
"epoch": 0.18629515632593552,
|
| 167 |
+
"grad_norm": 3.9677858352661133,
|
| 168 |
"learning_rate": 9.833333333333333e-06,
|
| 169 |
+
"loss": 0.5788,
|
| 170 |
"step": 575
|
| 171 |
},
|
| 172 |
{
|
| 173 |
+
"epoch": 0.19439494573141097,
|
| 174 |
+
"grad_norm": 3.9867312908172607,
|
| 175 |
"learning_rate": 9.777777777777779e-06,
|
| 176 |
+
"loss": 0.616,
|
| 177 |
"step": 600
|
| 178 |
},
|
| 179 |
{
|
| 180 |
+
"epoch": 0.20249473513688643,
|
| 181 |
+
"grad_norm": 4.993474960327148,
|
| 182 |
"learning_rate": 9.722222222222223e-06,
|
| 183 |
+
"loss": 0.6151,
|
| 184 |
"step": 625
|
| 185 |
},
|
| 186 |
{
|
| 187 |
+
"epoch": 0.2105945245423619,
|
| 188 |
+
"grad_norm": 4.049670219421387,
|
| 189 |
"learning_rate": 9.666666666666667e-06,
|
| 190 |
+
"loss": 0.5898,
|
| 191 |
"step": 650
|
| 192 |
},
|
| 193 |
{
|
| 194 |
+
"epoch": 0.21869431394783737,
|
| 195 |
+
"grad_norm": 3.711517095565796,
|
| 196 |
"learning_rate": 9.611111111111112e-06,
|
| 197 |
+
"loss": 0.5654,
|
| 198 |
"step": 675
|
| 199 |
},
|
| 200 |
{
|
| 201 |
+
"epoch": 0.22679410335331282,
|
| 202 |
+
"grad_norm": 4.081414222717285,
|
| 203 |
"learning_rate": 9.555555555555556e-06,
|
| 204 |
+
"loss": 0.5608,
|
| 205 |
"step": 700
|
| 206 |
},
|
| 207 |
{
|
| 208 |
+
"epoch": 0.23489389275878828,
|
| 209 |
+
"grad_norm": 4.10880708694458,
|
| 210 |
"learning_rate": 9.5e-06,
|
| 211 |
+
"loss": 0.5576,
|
| 212 |
"step": 725
|
| 213 |
},
|
| 214 |
{
|
| 215 |
+
"epoch": 0.24299368216426373,
|
| 216 |
+
"grad_norm": 3.764904737472534,
|
| 217 |
"learning_rate": 9.444444444444445e-06,
|
| 218 |
+
"loss": 0.5249,
|
| 219 |
"step": 750
|
| 220 |
},
|
| 221 |
{
|
| 222 |
+
"epoch": 0.2510934715697392,
|
| 223 |
+
"grad_norm": 3.7500803470611572,
|
| 224 |
"learning_rate": 9.38888888888889e-06,
|
| 225 |
+
"loss": 0.5724,
|
| 226 |
"step": 775
|
| 227 |
},
|
| 228 |
{
|
| 229 |
+
"epoch": 0.25919326097521467,
|
| 230 |
+
"grad_norm": 4.319125175476074,
|
| 231 |
"learning_rate": 9.333333333333334e-06,
|
| 232 |
+
"loss": 0.5368,
|
| 233 |
"step": 800
|
| 234 |
},
|
| 235 |
{
|
| 236 |
+
"epoch": 0.2672930503806901,
|
| 237 |
+
"grad_norm": 3.590097665786743,
|
| 238 |
"learning_rate": 9.277777777777778e-06,
|
| 239 |
+
"loss": 0.5494,
|
| 240 |
"step": 825
|
| 241 |
},
|
| 242 |
{
|
| 243 |
+
"epoch": 0.2753928397861656,
|
| 244 |
+
"grad_norm": 5.081935405731201,
|
| 245 |
"learning_rate": 9.222222222222224e-06,
|
| 246 |
+
"loss": 0.5443,
|
| 247 |
"step": 850
|
| 248 |
},
|
| 249 |
{
|
| 250 |
+
"epoch": 0.28349262919164103,
|
| 251 |
+
"grad_norm": 3.8603386878967285,
|
| 252 |
"learning_rate": 9.166666666666666e-06,
|
| 253 |
+
"loss": 0.5378,
|
| 254 |
"step": 875
|
| 255 |
},
|
| 256 |
{
|
| 257 |
+
"epoch": 0.2915924185971165,
|
| 258 |
+
"grad_norm": 3.7635509967803955,
|
| 259 |
"learning_rate": 9.111111111111112e-06,
|
| 260 |
+
"loss": 0.5121,
|
| 261 |
"step": 900
|
| 262 |
},
|
| 263 |
{
|
| 264 |
+
"epoch": 0.29969220800259194,
|
| 265 |
+
"grad_norm": 4.4741692543029785,
|
| 266 |
"learning_rate": 9.055555555555556e-06,
|
| 267 |
+
"loss": 0.5313,
|
| 268 |
"step": 925
|
| 269 |
},
|
| 270 |
{
|
| 271 |
+
"epoch": 0.3077919974080674,
|
| 272 |
+
"grad_norm": 4.126248836517334,
|
| 273 |
"learning_rate": 9e-06,
|
| 274 |
+
"loss": 0.5036,
|
| 275 |
"step": 950
|
| 276 |
},
|
| 277 |
{
|
| 278 |
+
"epoch": 0.31589178681354285,
|
| 279 |
+
"grad_norm": 3.2901930809020996,
|
| 280 |
"learning_rate": 8.944444444444446e-06,
|
| 281 |
+
"loss": 0.537,
|
| 282 |
"step": 975
|
| 283 |
},
|
| 284 |
{
|
| 285 |
+
"epoch": 0.3239915762190183,
|
| 286 |
+
"grad_norm": 4.152955532073975,
|
| 287 |
"learning_rate": 8.888888888888888e-06,
|
| 288 |
+
"loss": 0.5304,
|
| 289 |
"step": 1000
|
| 290 |
},
|
| 291 |
{
|
| 292 |
+
"epoch": 0.3239915762190183,
|
| 293 |
+
"eval_loss": 0.5235934853553772,
|
| 294 |
+
"eval_runtime": 2073.7226,
|
| 295 |
+
"eval_samples_per_second": 1.879,
|
| 296 |
+
"eval_steps_per_second": 0.118,
|
| 297 |
+
"eval_wer": 0.3785823799976412,
|
| 298 |
"step": 1000
|
| 299 |
},
|
| 300 |
{
|
| 301 |
+
"epoch": 0.33209136562449376,
|
| 302 |
+
"grad_norm": 3.5854649543762207,
|
| 303 |
"learning_rate": 8.833333333333334e-06,
|
| 304 |
+
"loss": 0.5265,
|
| 305 |
"step": 1025
|
| 306 |
},
|
| 307 |
{
|
| 308 |
+
"epoch": 0.3401911550299692,
|
| 309 |
+
"grad_norm": 3.5874733924865723,
|
| 310 |
"learning_rate": 8.777777777777778e-06,
|
| 311 |
+
"loss": 0.4957,
|
| 312 |
"step": 1050
|
| 313 |
},
|
| 314 |
{
|
| 315 |
+
"epoch": 0.34829094443544467,
|
| 316 |
+
"grad_norm": 4.2191338539123535,
|
| 317 |
"learning_rate": 8.722222222222224e-06,
|
| 318 |
+
"loss": 0.4892,
|
| 319 |
"step": 1075
|
| 320 |
},
|
| 321 |
{
|
| 322 |
+
"epoch": 0.3563907338409201,
|
| 323 |
+
"grad_norm": 4.080557823181152,
|
| 324 |
"learning_rate": 8.666666666666668e-06,
|
| 325 |
+
"loss": 0.4992,
|
| 326 |
"step": 1100
|
| 327 |
},
|
| 328 |
{
|
| 329 |
+
"epoch": 0.3644905232463956,
|
| 330 |
+
"grad_norm": 3.3749568462371826,
|
| 331 |
"learning_rate": 8.611111111111112e-06,
|
| 332 |
+
"loss": 0.4999,
|
| 333 |
"step": 1125
|
| 334 |
},
|
| 335 |
{
|
| 336 |
+
"epoch": 0.37259031265187104,
|
| 337 |
+
"grad_norm": 4.65897798538208,
|
| 338 |
"learning_rate": 8.555555555555556e-06,
|
| 339 |
+
"loss": 0.4893,
|
| 340 |
"step": 1150
|
| 341 |
},
|
| 342 |
{
|
| 343 |
+
"epoch": 0.3806901020573465,
|
| 344 |
+
"grad_norm": 4.345097541809082,
|
| 345 |
"learning_rate": 8.5e-06,
|
| 346 |
+
"loss": 0.5078,
|
| 347 |
"step": 1175
|
| 348 |
},
|
| 349 |
{
|
| 350 |
+
"epoch": 0.38878989146282195,
|
| 351 |
+
"grad_norm": 3.885808229446411,
|
| 352 |
"learning_rate": 8.444444444444446e-06,
|
| 353 |
+
"loss": 0.4683,
|
| 354 |
"step": 1200
|
| 355 |
},
|
| 356 |
{
|
| 357 |
+
"epoch": 0.3968896808682974,
|
| 358 |
+
"grad_norm": 3.44722318649292,
|
| 359 |
"learning_rate": 8.38888888888889e-06,
|
| 360 |
+
"loss": 0.488,
|
| 361 |
"step": 1225
|
| 362 |
},
|
| 363 |
{
|
| 364 |
+
"epoch": 0.40498947027377286,
|
| 365 |
+
"grad_norm": 3.6947743892669678,
|
| 366 |
"learning_rate": 8.333333333333334e-06,
|
| 367 |
+
"loss": 0.4743,
|
| 368 |
"step": 1250
|
| 369 |
},
|
| 370 |
{
|
| 371 |
+
"epoch": 0.41308925967924837,
|
| 372 |
+
"grad_norm": 4.338139533996582,
|
| 373 |
"learning_rate": 8.277777777777778e-06,
|
| 374 |
+
"loss": 0.4913,
|
| 375 |
"step": 1275
|
| 376 |
},
|
| 377 |
{
|
| 378 |
+
"epoch": 0.4211890490847238,
|
| 379 |
+
"grad_norm": 3.298166513442993,
|
| 380 |
"learning_rate": 8.222222222222222e-06,
|
| 381 |
+
"loss": 0.4865,
|
| 382 |
"step": 1300
|
| 383 |
},
|
| 384 |
{
|
| 385 |
+
"epoch": 0.4292888384901993,
|
| 386 |
+
"grad_norm": 3.604741334915161,
|
| 387 |
"learning_rate": 8.166666666666668e-06,
|
| 388 |
+
"loss": 0.4826,
|
| 389 |
"step": 1325
|
| 390 |
},
|
| 391 |
{
|
| 392 |
+
"epoch": 0.43738862789567473,
|
| 393 |
+
"grad_norm": 3.3499221801757812,
|
| 394 |
"learning_rate": 8.111111111111112e-06,
|
| 395 |
+
"loss": 0.4722,
|
| 396 |
"step": 1350
|
| 397 |
},
|
| 398 |
{
|
| 399 |
+
"epoch": 0.4454884173011502,
|
| 400 |
+
"grad_norm": 3.690180778503418,
|
| 401 |
"learning_rate": 8.055555555555557e-06,
|
| 402 |
+
"loss": 0.4911,
|
| 403 |
"step": 1375
|
| 404 |
},
|
| 405 |
{
|
| 406 |
+
"epoch": 0.45358820670662564,
|
| 407 |
+
"grad_norm": 2.948256492614746,
|
| 408 |
"learning_rate": 8.000000000000001e-06,
|
| 409 |
+
"loss": 0.4757,
|
| 410 |
"step": 1400
|
| 411 |
},
|
| 412 |
{
|
| 413 |
+
"epoch": 0.4616879961121011,
|
| 414 |
+
"grad_norm": 3.8048160076141357,
|
| 415 |
"learning_rate": 7.944444444444445e-06,
|
| 416 |
+
"loss": 0.4637,
|
| 417 |
"step": 1425
|
| 418 |
},
|
| 419 |
{
|
| 420 |
+
"epoch": 0.46978778551757655,
|
| 421 |
+
"grad_norm": 3.5490314960479736,
|
| 422 |
"learning_rate": 7.88888888888889e-06,
|
| 423 |
+
"loss": 0.4889,
|
| 424 |
"step": 1450
|
| 425 |
},
|
| 426 |
{
|
| 427 |
+
"epoch": 0.477887574923052,
|
| 428 |
+
"grad_norm": 5.1668009757995605,
|
| 429 |
"learning_rate": 7.833333333333333e-06,
|
| 430 |
+
"loss": 0.4595,
|
| 431 |
"step": 1475
|
| 432 |
},
|
| 433 |
{
|
| 434 |
+
"epoch": 0.48598736432852746,
|
| 435 |
+
"grad_norm": 3.1844429969787598,
|
| 436 |
"learning_rate": 7.77777777777778e-06,
|
| 437 |
+
"loss": 0.4799,
|
| 438 |
"step": 1500
|
| 439 |
},
|
| 440 |
{
|
| 441 |
+
"epoch": 0.4940871537340029,
|
| 442 |
+
"grad_norm": 3.055802583694458,
|
| 443 |
"learning_rate": 7.722222222222223e-06,
|
| 444 |
+
"loss": 0.4701,
|
| 445 |
"step": 1525
|
| 446 |
},
|
| 447 |
{
|
| 448 |
+
"epoch": 0.5021869431394784,
|
| 449 |
+
"grad_norm": 3.7342355251312256,
|
| 450 |
"learning_rate": 7.666666666666667e-06,
|
| 451 |
+
"loss": 0.4866,
|
| 452 |
"step": 1550
|
| 453 |
},
|
| 454 |
{
|
| 455 |
+
"epoch": 0.5102867325449538,
|
| 456 |
+
"grad_norm": 3.3651347160339355,
|
| 457 |
"learning_rate": 7.611111111111111e-06,
|
| 458 |
+
"loss": 0.4462,
|
| 459 |
"step": 1575
|
| 460 |
},
|
| 461 |
{
|
| 462 |
+
"epoch": 0.5183865219504293,
|
| 463 |
+
"grad_norm": 3.7382657527923584,
|
| 464 |
"learning_rate": 7.555555555555556e-06,
|
| 465 |
+
"loss": 0.4491,
|
| 466 |
"step": 1600
|
| 467 |
},
|
| 468 |
{
|
| 469 |
+
"epoch": 0.5264863113559047,
|
| 470 |
+
"grad_norm": 3.2548789978027344,
|
| 471 |
"learning_rate": 7.500000000000001e-06,
|
| 472 |
+
"loss": 0.4604,
|
| 473 |
"step": 1625
|
| 474 |
},
|
| 475 |
{
|
| 476 |
+
"epoch": 0.5345861007613802,
|
| 477 |
+
"grad_norm": 3.0430402755737305,
|
| 478 |
"learning_rate": 7.444444444444445e-06,
|
| 479 |
+
"loss": 0.4354,
|
| 480 |
"step": 1650
|
| 481 |
},
|
| 482 |
{
|
| 483 |
+
"epoch": 0.5426858901668556,
|
| 484 |
+
"grad_norm": 3.766554832458496,
|
| 485 |
"learning_rate": 7.38888888888889e-06,
|
| 486 |
+
"loss": 0.465,
|
| 487 |
"step": 1675
|
| 488 |
},
|
| 489 |
{
|
| 490 |
+
"epoch": 0.5507856795723312,
|
| 491 |
+
"grad_norm": 3.198700428009033,
|
| 492 |
"learning_rate": 7.333333333333333e-06,
|
| 493 |
+
"loss": 0.4695,
|
| 494 |
"step": 1700
|
| 495 |
},
|
| 496 |
{
|
| 497 |
+
"epoch": 0.5588854689778066,
|
| 498 |
+
"grad_norm": 2.7915539741516113,
|
| 499 |
"learning_rate": 7.277777777777778e-06,
|
| 500 |
+
"loss": 0.4511,
|
| 501 |
"step": 1725
|
| 502 |
},
|
| 503 |
{
|
| 504 |
+
"epoch": 0.5669852583832821,
|
| 505 |
+
"grad_norm": 3.3938117027282715,
|
| 506 |
"learning_rate": 7.222222222222223e-06,
|
| 507 |
+
"loss": 0.4266,
|
| 508 |
"step": 1750
|
| 509 |
},
|
| 510 |
{
|
| 511 |
+
"epoch": 0.5750850477887575,
|
| 512 |
+
"grad_norm": 3.4623613357543945,
|
| 513 |
"learning_rate": 7.166666666666667e-06,
|
| 514 |
+
"loss": 0.4483,
|
| 515 |
"step": 1775
|
| 516 |
},
|
| 517 |
{
|
| 518 |
+
"epoch": 0.583184837194233,
|
| 519 |
+
"grad_norm": 2.7853918075561523,
|
| 520 |
"learning_rate": 7.111111111111112e-06,
|
| 521 |
+
"loss": 0.4197,
|
| 522 |
"step": 1800
|
| 523 |
},
|
| 524 |
{
|
| 525 |
+
"epoch": 0.5912846265997084,
|
| 526 |
+
"grad_norm": 3.4827404022216797,
|
| 527 |
"learning_rate": 7.055555555555557e-06,
|
| 528 |
+
"loss": 0.4319,
|
| 529 |
"step": 1825
|
| 530 |
},
|
| 531 |
{
|
| 532 |
+
"epoch": 0.5993844160051839,
|
| 533 |
+
"grad_norm": 3.458853244781494,
|
| 534 |
"learning_rate": 7e-06,
|
| 535 |
+
"loss": 0.4301,
|
| 536 |
"step": 1850
|
| 537 |
},
|
| 538 |
{
|
| 539 |
+
"epoch": 0.6074842054106593,
|
| 540 |
+
"grad_norm": 3.726768970489502,
|
| 541 |
"learning_rate": 6.944444444444445e-06,
|
| 542 |
+
"loss": 0.4339,
|
| 543 |
"step": 1875
|
| 544 |
},
|
| 545 |
{
|
| 546 |
+
"epoch": 0.6155839948161348,
|
| 547 |
+
"grad_norm": 3.7597055435180664,
|
| 548 |
"learning_rate": 6.88888888888889e-06,
|
| 549 |
+
"loss": 0.4265,
|
| 550 |
"step": 1900
|
| 551 |
},
|
| 552 |
{
|
| 553 |
+
"epoch": 0.6236837842216102,
|
| 554 |
+
"grad_norm": 3.6740312576293945,
|
| 555 |
"learning_rate": 6.833333333333334e-06,
|
| 556 |
+
"loss": 0.4302,
|
| 557 |
"step": 1925
|
| 558 |
},
|
| 559 |
{
|
| 560 |
+
"epoch": 0.6317835736270857,
|
| 561 |
+
"grad_norm": 3.0535056591033936,
|
| 562 |
"learning_rate": 6.777777777777779e-06,
|
| 563 |
+
"loss": 0.4336,
|
| 564 |
"step": 1950
|
| 565 |
},
|
| 566 |
{
|
| 567 |
+
"epoch": 0.6398833630325611,
|
| 568 |
+
"grad_norm": 3.295497417449951,
|
| 569 |
"learning_rate": 6.7222222222222235e-06,
|
| 570 |
+
"loss": 0.4526,
|
| 571 |
"step": 1975
|
| 572 |
},
|
| 573 |
{
|
| 574 |
+
"epoch": 0.6479831524380366,
|
| 575 |
+
"grad_norm": 3.080634832382202,
|
| 576 |
"learning_rate": 6.666666666666667e-06,
|
| 577 |
+
"loss": 0.4409,
|
| 578 |
"step": 2000
|
| 579 |
},
|
| 580 |
{
|
| 581 |
+
"epoch": 0.6479831524380366,
|
| 582 |
+
"eval_loss": 0.4457571506500244,
|
| 583 |
+
"eval_runtime": 1994.8072,
|
| 584 |
+
"eval_samples_per_second": 1.954,
|
| 585 |
+
"eval_steps_per_second": 0.122,
|
| 586 |
+
"eval_wer": 0.35629201556787354,
|
| 587 |
"step": 2000
|
| 588 |
},
|
| 589 |
{
|
| 590 |
+
"epoch": 0.6560829418435121,
|
| 591 |
+
"grad_norm": 3.1193337440490723,
|
| 592 |
"learning_rate": 6.6111111111111115e-06,
|
| 593 |
+
"loss": 0.4111,
|
| 594 |
"step": 2025
|
| 595 |
},
|
| 596 |
{
|
| 597 |
+
"epoch": 0.6641827312489875,
|
| 598 |
+
"grad_norm": 3.554410934448242,
|
| 599 |
"learning_rate": 6.555555555555556e-06,
|
| 600 |
+
"loss": 0.4192,
|
| 601 |
"step": 2050
|
| 602 |
},
|
| 603 |
{
|
| 604 |
+
"epoch": 0.672282520654463,
|
| 605 |
+
"grad_norm": 4.468347549438477,
|
| 606 |
"learning_rate": 6.5000000000000004e-06,
|
| 607 |
+
"loss": 0.4329,
|
| 608 |
"step": 2075
|
| 609 |
},
|
| 610 |
{
|
| 611 |
+
"epoch": 0.6803823100599384,
|
| 612 |
+
"grad_norm": 3.5781023502349854,
|
| 613 |
"learning_rate": 6.444444444444445e-06,
|
| 614 |
+
"loss": 0.4191,
|
| 615 |
"step": 2100
|
| 616 |
},
|
| 617 |
{
|
| 618 |
+
"epoch": 0.6884820994654139,
|
| 619 |
+
"grad_norm": 3.257659912109375,
|
| 620 |
"learning_rate": 6.3888888888888885e-06,
|
| 621 |
+
"loss": 0.4046,
|
| 622 |
"step": 2125
|
| 623 |
},
|
| 624 |
{
|
| 625 |
+
"epoch": 0.6965818888708893,
|
| 626 |
+
"grad_norm": 3.030454397201538,
|
| 627 |
"learning_rate": 6.333333333333333e-06,
|
| 628 |
+
"loss": 0.4033,
|
| 629 |
"step": 2150
|
| 630 |
},
|
| 631 |
{
|
| 632 |
+
"epoch": 0.7046816782763649,
|
| 633 |
+
"grad_norm": 3.3289663791656494,
|
| 634 |
"learning_rate": 6.277777777777778e-06,
|
| 635 |
+
"loss": 0.4342,
|
| 636 |
"step": 2175
|
| 637 |
},
|
| 638 |
{
|
| 639 |
+
"epoch": 0.7127814676818403,
|
| 640 |
+
"grad_norm": 3.4743051528930664,
|
| 641 |
"learning_rate": 6.222222222222223e-06,
|
| 642 |
+
"loss": 0.4102,
|
| 643 |
"step": 2200
|
| 644 |
},
|
| 645 |
{
|
| 646 |
+
"epoch": 0.7208812570873158,
|
| 647 |
+
"grad_norm": 3.300485610961914,
|
| 648 |
"learning_rate": 6.166666666666667e-06,
|
| 649 |
+
"loss": 0.3951,
|
| 650 |
"step": 2225
|
| 651 |
},
|
| 652 |
{
|
| 653 |
+
"epoch": 0.7289810464927912,
|
| 654 |
+
"grad_norm": 3.374356746673584,
|
| 655 |
"learning_rate": 6.111111111111112e-06,
|
| 656 |
+
"loss": 0.3993,
|
| 657 |
"step": 2250
|
| 658 |
},
|
| 659 |
{
|
| 660 |
+
"epoch": 0.7370808358982667,
|
| 661 |
+
"grad_norm": 3.108508586883545,
|
| 662 |
"learning_rate": 6.055555555555555e-06,
|
| 663 |
+
"loss": 0.3927,
|
| 664 |
"step": 2275
|
| 665 |
},
|
| 666 |
{
|
| 667 |
+
"epoch": 0.7451806253037421,
|
| 668 |
+
"grad_norm": 3.448174238204956,
|
| 669 |
"learning_rate": 6e-06,
|
| 670 |
+
"loss": 0.4254,
|
| 671 |
"step": 2300
|
| 672 |
},
|
| 673 |
{
|
| 674 |
+
"epoch": 0.7532804147092176,
|
| 675 |
+
"grad_norm": 3.17012095451355,
|
| 676 |
"learning_rate": 5.944444444444445e-06,
|
| 677 |
+
"loss": 0.396,
|
| 678 |
"step": 2325
|
| 679 |
},
|
| 680 |
{
|
| 681 |
+
"epoch": 0.761380204114693,
|
| 682 |
+
"grad_norm": 3.3391034603118896,
|
| 683 |
"learning_rate": 5.88888888888889e-06,
|
| 684 |
+
"loss": 0.4035,
|
| 685 |
"step": 2350
|
| 686 |
},
|
| 687 |
{
|
| 688 |
+
"epoch": 0.7694799935201685,
|
| 689 |
+
"grad_norm": 3.1390321254730225,
|
| 690 |
"learning_rate": 5.833333333333334e-06,
|
| 691 |
+
"loss": 0.4387,
|
| 692 |
"step": 2375
|
| 693 |
},
|
| 694 |
{
|
| 695 |
+
"epoch": 0.7775797829256439,
|
| 696 |
+
"grad_norm": 2.801060199737549,
|
| 697 |
+
"learning_rate": 5.78e-06,
|
| 698 |
+
"loss": 0.4101,
|
| 699 |
"step": 2400
|
| 700 |
},
|
| 701 |
{
|
| 702 |
+
"epoch": 0.7856795723311194,
|
| 703 |
+
"grad_norm": 3.169654130935669,
|
| 704 |
+
"learning_rate": 5.724444444444445e-06,
|
| 705 |
+
"loss": 0.4019,
|
| 706 |
"step": 2425
|
| 707 |
},
|
| 708 |
{
|
| 709 |
+
"epoch": 0.7937793617365948,
|
| 710 |
+
"grad_norm": 3.1517333984375,
|
| 711 |
+
"learning_rate": 5.6688888888888895e-06,
|
| 712 |
+
"loss": 0.3955,
|
| 713 |
"step": 2450
|
| 714 |
},
|
| 715 |
{
|
| 716 |
+
"epoch": 0.8018791511420703,
|
| 717 |
+
"grad_norm": 2.812309980392456,
|
| 718 |
+
"learning_rate": 5.613333333333334e-06,
|
| 719 |
+
"loss": 0.3976,
|
| 720 |
"step": 2475
|
| 721 |
},
|
| 722 |
{
|
| 723 |
+
"epoch": 0.8099789405475457,
|
| 724 |
+
"grad_norm": 3.095435619354248,
|
| 725 |
+
"learning_rate": 5.557777777777778e-06,
|
| 726 |
+
"loss": 0.4173,
|
| 727 |
"step": 2500
|
| 728 |
},
|
| 729 |
{
|
| 730 |
+
"epoch": 0.8180787299530212,
|
| 731 |
+
"grad_norm": 3.530505418777466,
|
| 732 |
+
"learning_rate": 5.5022222222222224e-06,
|
| 733 |
+
"loss": 0.3968,
|
| 734 |
"step": 2525
|
| 735 |
},
|
| 736 |
{
|
| 737 |
+
"epoch": 0.8261785193584967,
|
| 738 |
+
"grad_norm": 3.6718244552612305,
|
| 739 |
+
"learning_rate": 5.4466666666666665e-06,
|
| 740 |
+
"loss": 0.4365,
|
| 741 |
"step": 2550
|
| 742 |
},
|
| 743 |
{
|
| 744 |
+
"epoch": 0.8342783087639721,
|
| 745 |
+
"grad_norm": 3.455793857574463,
|
| 746 |
+
"learning_rate": 5.391111111111111e-06,
|
| 747 |
+
"loss": 0.3995,
|
| 748 |
"step": 2575
|
| 749 |
},
|
| 750 |
{
|
| 751 |
+
"epoch": 0.8423780981694476,
|
| 752 |
+
"grad_norm": 3.074331760406494,
|
| 753 |
+
"learning_rate": 5.335555555555556e-06,
|
| 754 |
+
"loss": 0.3995,
|
| 755 |
"step": 2600
|
| 756 |
},
|
| 757 |
{
|
| 758 |
+
"epoch": 0.850477887574923,
|
| 759 |
+
"grad_norm": 3.732499361038208,
|
| 760 |
+
"learning_rate": 5.28e-06,
|
| 761 |
+
"loss": 0.3873,
|
| 762 |
"step": 2625
|
| 763 |
},
|
| 764 |
{
|
| 765 |
+
"epoch": 0.8585776769803986,
|
| 766 |
+
"grad_norm": 2.719492197036743,
|
| 767 |
+
"learning_rate": 5.224444444444445e-06,
|
| 768 |
+
"loss": 0.3938,
|
| 769 |
"step": 2650
|
| 770 |
},
|
| 771 |
{
|
| 772 |
+
"epoch": 0.866677466385874,
|
| 773 |
+
"grad_norm": 3.279792308807373,
|
| 774 |
+
"learning_rate": 5.168888888888889e-06,
|
| 775 |
+
"loss": 0.4087,
|
| 776 |
"step": 2675
|
| 777 |
},
|
| 778 |
{
|
| 779 |
+
"epoch": 0.8747772557913495,
|
| 780 |
+
"grad_norm": 3.178786516189575,
|
| 781 |
+
"learning_rate": 5.113333333333333e-06,
|
| 782 |
+
"loss": 0.3771,
|
| 783 |
"step": 2700
|
| 784 |
},
|
| 785 |
{
|
| 786 |
+
"epoch": 0.8828770451968249,
|
| 787 |
+
"grad_norm": 2.70697283744812,
|
| 788 |
+
"learning_rate": 5.057777777777778e-06,
|
| 789 |
+
"loss": 0.3991,
|
| 790 |
"step": 2725
|
| 791 |
},
|
| 792 |
{
|
| 793 |
+
"epoch": 0.8909768346023004,
|
| 794 |
+
"grad_norm": 3.0300474166870117,
|
| 795 |
+
"learning_rate": 5.002222222222223e-06,
|
| 796 |
+
"loss": 0.4069,
|
| 797 |
"step": 2750
|
| 798 |
},
|
| 799 |
{
|
| 800 |
+
"epoch": 0.8990766240077758,
|
| 801 |
+
"grad_norm": 4.001626491546631,
|
| 802 |
+
"learning_rate": 4.946666666666667e-06,
|
| 803 |
+
"loss": 0.4153,
|
| 804 |
"step": 2775
|
| 805 |
},
|
| 806 |
{
|
| 807 |
+
"epoch": 0.9071764134132513,
|
| 808 |
+
"grad_norm": 3.1544246673583984,
|
| 809 |
+
"learning_rate": 4.891111111111111e-06,
|
| 810 |
+
"loss": 0.3733,
|
| 811 |
"step": 2800
|
| 812 |
},
|
| 813 |
{
|
| 814 |
+
"epoch": 0.9152762028187267,
|
| 815 |
+
"grad_norm": 3.553840398788452,
|
| 816 |
+
"learning_rate": 4.835555555555556e-06,
|
| 817 |
+
"loss": 0.409,
|
| 818 |
"step": 2825
|
| 819 |
},
|
| 820 |
{
|
| 821 |
+
"epoch": 0.9233759922242022,
|
| 822 |
+
"grad_norm": 3.6127443313598633,
|
| 823 |
+
"learning_rate": 4.78e-06,
|
| 824 |
+
"loss": 0.3728,
|
| 825 |
"step": 2850
|
| 826 |
},
|
| 827 |
{
|
| 828 |
+
"epoch": 0.9314757816296776,
|
| 829 |
+
"grad_norm": 3.1611216068267822,
|
| 830 |
+
"learning_rate": 4.724444444444445e-06,
|
| 831 |
+
"loss": 0.3923,
|
| 832 |
"step": 2875
|
| 833 |
},
|
| 834 |
{
|
| 835 |
+
"epoch": 0.9395755710351531,
|
| 836 |
+
"grad_norm": 3.2189152240753174,
|
| 837 |
+
"learning_rate": 4.66888888888889e-06,
|
| 838 |
+
"loss": 0.4179,
|
| 839 |
"step": 2900
|
| 840 |
},
|
| 841 |
{
|
| 842 |
+
"epoch": 0.9476753604406285,
|
| 843 |
+
"grad_norm": 3.3200769424438477,
|
| 844 |
+
"learning_rate": 4.613333333333334e-06,
|
| 845 |
+
"loss": 0.4067,
|
| 846 |
"step": 2925
|
| 847 |
},
|
| 848 |
{
|
| 849 |
+
"epoch": 0.955775149846104,
|
| 850 |
+
"grad_norm": 3.38618540763855,
|
| 851 |
+
"learning_rate": 4.557777777777778e-06,
|
| 852 |
+
"loss": 0.384,
|
| 853 |
"step": 2950
|
| 854 |
},
|
| 855 |
{
|
| 856 |
+
"epoch": 0.9638749392515794,
|
| 857 |
+
"grad_norm": 3.5702147483825684,
|
| 858 |
+
"learning_rate": 4.502222222222223e-06,
|
| 859 |
+
"loss": 0.3949,
|
| 860 |
"step": 2975
|
| 861 |
},
|
| 862 |
{
|
| 863 |
+
"epoch": 0.9719747286570549,
|
| 864 |
+
"grad_norm": 2.6733343601226807,
|
| 865 |
+
"learning_rate": 4.446666666666667e-06,
|
| 866 |
+
"loss": 0.3914,
|
| 867 |
"step": 3000
|
| 868 |
},
|
| 869 |
{
|
| 870 |
+
"epoch": 0.9719747286570549,
|
| 871 |
+
"eval_loss": 0.40348634123802185,
|
| 872 |
+
"eval_runtime": 1837.3277,
|
| 873 |
+
"eval_samples_per_second": 2.121,
|
| 874 |
+
"eval_steps_per_second": 0.133,
|
| 875 |
+
"eval_wer": 0.3089751149899752,
|
| 876 |
"step": 3000
|
| 877 |
},
|
| 878 |
{
|
| 879 |
+
"epoch": 0.9800745180625303,
|
| 880 |
+
"grad_norm": 3.1163229942321777,
|
| 881 |
+
"learning_rate": 4.391111111111112e-06,
|
| 882 |
+
"loss": 0.3614,
|
| 883 |
"step": 3025
|
| 884 |
},
|
| 885 |
{
|
| 886 |
+
"epoch": 0.9881743074680058,
|
| 887 |
+
"grad_norm": 4.919127941131592,
|
| 888 |
+
"learning_rate": 4.3355555555555565e-06,
|
| 889 |
+
"loss": 0.3815,
|
| 890 |
"step": 3050
|
| 891 |
},
|
| 892 |
{
|
| 893 |
+
"epoch": 0.9962740968734813,
|
| 894 |
+
"grad_norm": 3.523918867111206,
|
| 895 |
+
"learning_rate": 4.2800000000000005e-06,
|
| 896 |
+
"loss": 0.3679,
|
| 897 |
"step": 3075
|
| 898 |
},
|
| 899 |
{
|
| 900 |
+
"epoch": 1.0042118904908472,
|
| 901 |
+
"grad_norm": 2.188014268875122,
|
| 902 |
+
"learning_rate": 4.2244444444444446e-06,
|
| 903 |
+
"loss": 0.3305,
|
| 904 |
"step": 3100
|
| 905 |
},
|
| 906 |
{
|
| 907 |
+
"epoch": 1.0123116798963228,
|
| 908 |
+
"grad_norm": 3.0015525817871094,
|
| 909 |
+
"learning_rate": 4.168888888888889e-06,
|
| 910 |
+
"loss": 0.2856,
|
| 911 |
"step": 3125
|
| 912 |
},
|
| 913 |
{
|
| 914 |
+
"epoch": 1.020411469301798,
|
| 915 |
+
"grad_norm": 3.565284490585327,
|
| 916 |
+
"learning_rate": 4.1133333333333335e-06,
|
| 917 |
+
"loss": 0.296,
|
| 918 |
"step": 3150
|
| 919 |
},
|
| 920 |
{
|
| 921 |
+
"epoch": 1.0285112587072736,
|
| 922 |
+
"grad_norm": 2.8771865367889404,
|
| 923 |
+
"learning_rate": 4.057777777777778e-06,
|
| 924 |
+
"loss": 0.2912,
|
| 925 |
"step": 3175
|
| 926 |
},
|
| 927 |
{
|
| 928 |
+
"epoch": 1.036611048112749,
|
| 929 |
+
"grad_norm": 2.7181403636932373,
|
| 930 |
+
"learning_rate": 4.002222222222222e-06,
|
| 931 |
+
"loss": 0.2871,
|
| 932 |
"step": 3200
|
| 933 |
},
|
| 934 |
{
|
| 935 |
+
"epoch": 1.0447108375182246,
|
| 936 |
+
"grad_norm": 2.652308225631714,
|
| 937 |
+
"learning_rate": 3.946666666666667e-06,
|
| 938 |
+
"loss": 0.2669,
|
| 939 |
"step": 3225
|
| 940 |
},
|
| 941 |
{
|
| 942 |
+
"epoch": 1.0528106269237,
|
| 943 |
+
"grad_norm": 2.390458345413208,
|
| 944 |
+
"learning_rate": 3.891111111111111e-06,
|
| 945 |
+
"loss": 0.2852,
|
| 946 |
"step": 3250
|
| 947 |
},
|
| 948 |
{
|
| 949 |
+
"epoch": 1.0609104163291754,
|
| 950 |
+
"grad_norm": 3.036996603012085,
|
| 951 |
+
"learning_rate": 3.835555555555555e-06,
|
| 952 |
+
"loss": 0.2835,
|
| 953 |
"step": 3275
|
| 954 |
},
|
| 955 |
{
|
| 956 |
+
"epoch": 1.0690102057346509,
|
| 957 |
+
"grad_norm": 3.0076892375946045,
|
| 958 |
+
"learning_rate": 3.7800000000000002e-06,
|
| 959 |
+
"loss": 0.2853,
|
| 960 |
"step": 3300
|
| 961 |
},
|
| 962 |
{
|
| 963 |
+
"epoch": 1.0771099951401264,
|
| 964 |
+
"grad_norm": 2.6001665592193604,
|
| 965 |
+
"learning_rate": 3.724444444444445e-06,
|
| 966 |
+
"loss": 0.2552,
|
| 967 |
"step": 3325
|
| 968 |
},
|
| 969 |
{
|
| 970 |
+
"epoch": 1.085209784545602,
|
| 971 |
+
"grad_norm": 2.9106733798980713,
|
| 972 |
+
"learning_rate": 3.668888888888889e-06,
|
| 973 |
+
"loss": 0.2921,
|
| 974 |
"step": 3350
|
| 975 |
},
|
| 976 |
{
|
| 977 |
+
"epoch": 1.0933095739510772,
|
| 978 |
+
"grad_norm": 2.579561710357666,
|
| 979 |
+
"learning_rate": 3.6133333333333336e-06,
|
| 980 |
+
"loss": 0.2904,
|
| 981 |
"step": 3375
|
| 982 |
},
|
| 983 |
{
|
| 984 |
+
"epoch": 1.1014093633565527,
|
| 985 |
+
"grad_norm": 2.4961764812469482,
|
| 986 |
+
"learning_rate": 3.5577777777777785e-06,
|
| 987 |
+
"loss": 0.2699,
|
| 988 |
"step": 3400
|
| 989 |
},
|
| 990 |
{
|
| 991 |
+
"epoch": 1.1095091527620282,
|
| 992 |
+
"grad_norm": 2.9800219535827637,
|
| 993 |
+
"learning_rate": 3.5022222222222225e-06,
|
| 994 |
+
"loss": 0.2867,
|
| 995 |
"step": 3425
|
| 996 |
},
|
| 997 |
{
|
| 998 |
+
"epoch": 1.1176089421675037,
|
| 999 |
+
"grad_norm": 2.7929482460021973,
|
| 1000 |
+
"learning_rate": 3.446666666666667e-06,
|
| 1001 |
+
"loss": 0.2707,
|
| 1002 |
"step": 3450
|
| 1003 |
},
|
| 1004 |
{
|
| 1005 |
+
"epoch": 1.125708731572979,
|
| 1006 |
+
"grad_norm": 3.021549940109253,
|
| 1007 |
+
"learning_rate": 3.391111111111111e-06,
|
| 1008 |
+
"loss": 0.2793,
|
| 1009 |
"step": 3475
|
| 1010 |
},
|
| 1011 |
{
|
| 1012 |
+
"epoch": 1.1338085209784545,
|
| 1013 |
+
"grad_norm": 2.6613991260528564,
|
| 1014 |
+
"learning_rate": 3.335555555555556e-06,
|
| 1015 |
+
"loss": 0.2705,
|
| 1016 |
"step": 3500
|
| 1017 |
},
|
| 1018 |
{
|
| 1019 |
+
"epoch": 1.14190831038393,
|
| 1020 |
+
"grad_norm": 2.4502217769622803,
|
| 1021 |
+
"learning_rate": 3.2800000000000004e-06,
|
| 1022 |
+
"loss": 0.2776,
|
| 1023 |
"step": 3525
|
| 1024 |
},
|
| 1025 |
{
|
| 1026 |
+
"epoch": 1.1500080997894055,
|
| 1027 |
+
"grad_norm": 2.698561668395996,
|
| 1028 |
+
"learning_rate": 3.2244444444444444e-06,
|
| 1029 |
+
"loss": 0.2685,
|
| 1030 |
"step": 3550
|
| 1031 |
},
|
| 1032 |
{
|
| 1033 |
+
"epoch": 1.158107889194881,
|
| 1034 |
+
"grad_norm": 2.5799190998077393,
|
| 1035 |
+
"learning_rate": 3.1688888888888893e-06,
|
| 1036 |
+
"loss": 0.2788,
|
| 1037 |
"step": 3575
|
| 1038 |
},
|
| 1039 |
{
|
| 1040 |
+
"epoch": 1.1662076786003563,
|
| 1041 |
+
"grad_norm": 2.5856029987335205,
|
| 1042 |
+
"learning_rate": 3.1133333333333337e-06,
|
| 1043 |
+
"loss": 0.2691,
|
| 1044 |
"step": 3600
|
| 1045 |
},
|
| 1046 |
{
|
| 1047 |
+
"epoch": 1.1743074680058319,
|
| 1048 |
+
"grad_norm": 2.861074209213257,
|
| 1049 |
+
"learning_rate": 3.0577777777777778e-06,
|
| 1050 |
+
"loss": 0.2862,
|
| 1051 |
"step": 3625
|
| 1052 |
},
|
| 1053 |
{
|
| 1054 |
+
"epoch": 1.1824072574113074,
|
| 1055 |
+
"grad_norm": 2.6794049739837646,
|
| 1056 |
+
"learning_rate": 3.0022222222222227e-06,
|
| 1057 |
+
"loss": 0.269,
|
| 1058 |
"step": 3650
|
| 1059 |
},
|
| 1060 |
{
|
| 1061 |
+
"epoch": 1.1905070468167827,
|
| 1062 |
+
"grad_norm": 2.6813509464263916,
|
| 1063 |
+
"learning_rate": 2.946666666666667e-06,
|
| 1064 |
+
"loss": 0.2787,
|
| 1065 |
"step": 3675
|
| 1066 |
},
|
| 1067 |
{
|
| 1068 |
+
"epoch": 1.1986068362222582,
|
| 1069 |
+
"grad_norm": 2.5139195919036865,
|
| 1070 |
+
"learning_rate": 2.891111111111111e-06,
|
| 1071 |
+
"loss": 0.2876,
|
| 1072 |
"step": 3700
|
| 1073 |
},
|
| 1074 |
{
|
| 1075 |
+
"epoch": 1.2067066256277337,
|
| 1076 |
+
"grad_norm": 2.5221784114837646,
|
| 1077 |
+
"learning_rate": 2.835555555555556e-06,
|
| 1078 |
+
"loss": 0.2629,
|
| 1079 |
"step": 3725
|
| 1080 |
},
|
| 1081 |
{
|
| 1082 |
+
"epoch": 1.2148064150332092,
|
| 1083 |
+
"grad_norm": 2.6317663192749023,
|
| 1084 |
+
"learning_rate": 2.7800000000000005e-06,
|
| 1085 |
+
"loss": 0.2792,
|
| 1086 |
"step": 3750
|
| 1087 |
},
|
| 1088 |
{
|
| 1089 |
+
"epoch": 1.2229062044386847,
|
| 1090 |
+
"grad_norm": 3.165544271469116,
|
| 1091 |
+
"learning_rate": 2.7244444444444445e-06,
|
| 1092 |
+
"loss": 0.2798,
|
| 1093 |
"step": 3775
|
| 1094 |
},
|
| 1095 |
{
|
| 1096 |
+
"epoch": 1.23100599384416,
|
| 1097 |
+
"grad_norm": 2.9240896701812744,
|
| 1098 |
+
"learning_rate": 2.6688888888888894e-06,
|
| 1099 |
+
"loss": 0.2745,
|
| 1100 |
"step": 3800
|
| 1101 |
},
|
| 1102 |
{
|
| 1103 |
+
"epoch": 1.2391057832496355,
|
| 1104 |
+
"grad_norm": 2.7815427780151367,
|
| 1105 |
+
"learning_rate": 2.6133333333333334e-06,
|
| 1106 |
+
"loss": 0.2768,
|
| 1107 |
"step": 3825
|
| 1108 |
},
|
| 1109 |
{
|
| 1110 |
+
"epoch": 1.247205572655111,
|
| 1111 |
+
"grad_norm": 3.016592025756836,
|
| 1112 |
+
"learning_rate": 2.557777777777778e-06,
|
| 1113 |
+
"loss": 0.2735,
|
| 1114 |
"step": 3850
|
| 1115 |
},
|
| 1116 |
{
|
| 1117 |
+
"epoch": 1.2553053620605863,
|
| 1118 |
+
"grad_norm": 2.7359392642974854,
|
| 1119 |
+
"learning_rate": 2.5022222222222224e-06,
|
| 1120 |
+
"loss": 0.2757,
|
| 1121 |
"step": 3875
|
| 1122 |
},
|
| 1123 |
{
|
| 1124 |
+
"epoch": 1.2634051514660618,
|
| 1125 |
+
"grad_norm": 2.575500726699829,
|
| 1126 |
+
"learning_rate": 2.446666666666667e-06,
|
| 1127 |
+
"loss": 0.2635,
|
| 1128 |
"step": 3900
|
| 1129 |
},
|
| 1130 |
{
|
| 1131 |
+
"epoch": 1.2715049408715373,
|
| 1132 |
+
"grad_norm": 2.202298164367676,
|
| 1133 |
+
"learning_rate": 2.3911111111111113e-06,
|
| 1134 |
+
"loss": 0.2682,
|
| 1135 |
"step": 3925
|
| 1136 |
},
|
| 1137 |
{
|
| 1138 |
+
"epoch": 1.2796047302770128,
|
| 1139 |
+
"grad_norm": 3.1898550987243652,
|
| 1140 |
+
"learning_rate": 2.3355555555555557e-06,
|
| 1141 |
+
"loss": 0.2708,
|
| 1142 |
"step": 3950
|
| 1143 |
},
|
| 1144 |
{
|
| 1145 |
+
"epoch": 1.2877045196824883,
|
| 1146 |
+
"grad_norm": 2.5168306827545166,
|
| 1147 |
+
"learning_rate": 2.28e-06,
|
| 1148 |
+
"loss": 0.2639,
|
| 1149 |
"step": 3975
|
| 1150 |
},
|
| 1151 |
{
|
| 1152 |
+
"epoch": 1.2958043090879636,
|
| 1153 |
+
"grad_norm": 2.682749032974243,
|
| 1154 |
+
"learning_rate": 2.2244444444444447e-06,
|
| 1155 |
+
"loss": 0.296,
|
| 1156 |
"step": 4000
|
| 1157 |
},
|
| 1158 |
{
|
| 1159 |
+
"epoch": 1.2958043090879636,
|
| 1160 |
+
"eval_loss": 0.3868160843849182,
|
| 1161 |
+
"eval_runtime": 1836.049,
|
| 1162 |
+
"eval_samples_per_second": 2.122,
|
| 1163 |
+
"eval_steps_per_second": 0.133,
|
| 1164 |
+
"eval_wer": 0.2977002004953414,
|
| 1165 |
"step": 4000
|
| 1166 |
},
|
| 1167 |
{
|
| 1168 |
+
"epoch": 1.3039040984934391,
|
| 1169 |
+
"grad_norm": 2.805448293685913,
|
| 1170 |
+
"learning_rate": 2.168888888888889e-06,
|
| 1171 |
+
"loss": 0.2728,
|
| 1172 |
"step": 4025
|
| 1173 |
},
|
| 1174 |
{
|
| 1175 |
+
"epoch": 1.3120038878989146,
|
| 1176 |
+
"grad_norm": 2.2613837718963623,
|
| 1177 |
+
"learning_rate": 2.1133333333333336e-06,
|
| 1178 |
+
"loss": 0.2695,
|
| 1179 |
"step": 4050
|
| 1180 |
},
|
| 1181 |
{
|
| 1182 |
+
"epoch": 1.3201036773043902,
|
| 1183 |
+
"grad_norm": 2.4388859272003174,
|
| 1184 |
+
"learning_rate": 2.057777777777778e-06,
|
| 1185 |
+
"loss": 0.286,
|
| 1186 |
"step": 4075
|
| 1187 |
},
|
| 1188 |
{
|
| 1189 |
+
"epoch": 1.3282034667098657,
|
| 1190 |
+
"grad_norm": 2.807706832885742,
|
| 1191 |
+
"learning_rate": 2.0022222222222225e-06,
|
| 1192 |
+
"loss": 0.251,
|
| 1193 |
"step": 4100
|
| 1194 |
},
|
| 1195 |
{
|
| 1196 |
+
"epoch": 1.336303256115341,
|
| 1197 |
+
"grad_norm": 2.927849292755127,
|
| 1198 |
+
"learning_rate": 1.9466666666666665e-06,
|
| 1199 |
+
"loss": 0.2747,
|
| 1200 |
"step": 4125
|
| 1201 |
},
|
| 1202 |
{
|
| 1203 |
+
"epoch": 1.3444030455208165,
|
| 1204 |
+
"grad_norm": 2.2061877250671387,
|
| 1205 |
+
"learning_rate": 1.8911111111111114e-06,
|
| 1206 |
+
"loss": 0.267,
|
| 1207 |
"step": 4150
|
| 1208 |
},
|
| 1209 |
{
|
| 1210 |
+
"epoch": 1.352502834926292,
|
| 1211 |
+
"grad_norm": 2.5290238857269287,
|
| 1212 |
+
"learning_rate": 1.8355555555555557e-06,
|
| 1213 |
+
"loss": 0.2727,
|
| 1214 |
"step": 4175
|
| 1215 |
},
|
| 1216 |
{
|
| 1217 |
+
"epoch": 1.3606026243317673,
|
| 1218 |
+
"grad_norm": 2.74303936958313,
|
| 1219 |
+
"learning_rate": 1.7800000000000001e-06,
|
| 1220 |
+
"loss": 0.2726,
|
| 1221 |
"step": 4200
|
| 1222 |
},
|
| 1223 |
{
|
| 1224 |
+
"epoch": 1.3687024137372428,
|
| 1225 |
+
"grad_norm": 2.194861888885498,
|
| 1226 |
+
"learning_rate": 1.7244444444444448e-06,
|
| 1227 |
+
"loss": 0.2645,
|
| 1228 |
"step": 4225
|
| 1229 |
},
|
| 1230 |
{
|
| 1231 |
+
"epoch": 1.3768022031427183,
|
| 1232 |
+
"grad_norm": 2.4960832595825195,
|
| 1233 |
+
"learning_rate": 1.668888888888889e-06,
|
| 1234 |
+
"loss": 0.2788,
|
| 1235 |
"step": 4250
|
| 1236 |
},
|
| 1237 |
{
|
| 1238 |
+
"epoch": 1.3849019925481938,
|
| 1239 |
+
"grad_norm": 2.3262386322021484,
|
| 1240 |
+
"learning_rate": 1.6133333333333335e-06,
|
| 1241 |
+
"loss": 0.2586,
|
| 1242 |
"step": 4275
|
| 1243 |
},
|
| 1244 |
{
|
| 1245 |
+
"epoch": 1.3930017819536693,
|
| 1246 |
+
"grad_norm": 2.5423128604888916,
|
| 1247 |
+
"learning_rate": 1.5577777777777777e-06,
|
| 1248 |
+
"loss": 0.2719,
|
| 1249 |
"step": 4300
|
| 1250 |
},
|
| 1251 |
{
|
| 1252 |
+
"epoch": 1.4011015713591446,
|
| 1253 |
+
"grad_norm": 2.9188232421875,
|
| 1254 |
+
"learning_rate": 1.5022222222222224e-06,
|
| 1255 |
+
"loss": 0.2671,
|
| 1256 |
"step": 4325
|
| 1257 |
},
|
| 1258 |
{
|
| 1259 |
+
"epoch": 1.40920136076462,
|
| 1260 |
+
"grad_norm": 2.624691963195801,
|
| 1261 |
+
"learning_rate": 1.4466666666666669e-06,
|
| 1262 |
+
"loss": 0.2809,
|
| 1263 |
"step": 4350
|
| 1264 |
},
|
| 1265 |
{
|
| 1266 |
+
"epoch": 1.4173011501700956,
|
| 1267 |
+
"grad_norm": 2.674189329147339,
|
| 1268 |
+
"learning_rate": 1.3911111111111111e-06,
|
| 1269 |
+
"loss": 0.2651,
|
| 1270 |
"step": 4375
|
| 1271 |
},
|
| 1272 |
{
|
| 1273 |
+
"epoch": 1.425400939575571,
|
| 1274 |
+
"grad_norm": 2.832871913909912,
|
| 1275 |
+
"learning_rate": 1.3355555555555558e-06,
|
| 1276 |
+
"loss": 0.2676,
|
| 1277 |
"step": 4400
|
| 1278 |
},
|
| 1279 |
{
|
| 1280 |
+
"epoch": 1.4335007289810464,
|
| 1281 |
+
"grad_norm": 2.8886163234710693,
|
| 1282 |
"learning_rate": 1.28e-06,
|
| 1283 |
+
"loss": 0.2717,
|
| 1284 |
"step": 4425
|
| 1285 |
},
|
| 1286 |
{
|
| 1287 |
+
"epoch": 1.441600518386522,
|
| 1288 |
+
"grad_norm": 2.9759998321533203,
|
| 1289 |
"learning_rate": 1.2244444444444445e-06,
|
| 1290 |
+
"loss": 0.2709,
|
| 1291 |
"step": 4450
|
| 1292 |
},
|
| 1293 |
{
|
| 1294 |
+
"epoch": 1.4497003077919974,
|
| 1295 |
+
"grad_norm": 2.8396642208099365,
|
| 1296 |
"learning_rate": 1.168888888888889e-06,
|
| 1297 |
+
"loss": 0.2701,
|
| 1298 |
"step": 4475
|
| 1299 |
},
|
| 1300 |
{
|
| 1301 |
+
"epoch": 1.457800097197473,
|
| 1302 |
+
"grad_norm": 3.0574100017547607,
|
| 1303 |
+
"learning_rate": 1.1155555555555558e-06,
|
| 1304 |
+
"loss": 0.2592,
|
| 1305 |
"step": 4500
|
| 1306 |
},
|
| 1307 |
{
|
| 1308 |
+
"epoch": 1.4658998866029482,
|
| 1309 |
+
"grad_norm": 2.318142890930176,
|
| 1310 |
+
"learning_rate": 1.06e-06,
|
| 1311 |
+
"loss": 0.2731,
|
| 1312 |
"step": 4525
|
| 1313 |
},
|
| 1314 |
{
|
| 1315 |
+
"epoch": 1.4739996760084237,
|
| 1316 |
+
"grad_norm": 2.372235059738159,
|
| 1317 |
+
"learning_rate": 1.0044444444444445e-06,
|
| 1318 |
+
"loss": 0.2694,
|
| 1319 |
"step": 4550
|
| 1320 |
},
|
| 1321 |
{
|
| 1322 |
+
"epoch": 1.4820994654138993,
|
| 1323 |
+
"grad_norm": 2.6232292652130127,
|
| 1324 |
+
"learning_rate": 9.488888888888889e-07,
|
| 1325 |
+
"loss": 0.2529,
|
| 1326 |
"step": 4575
|
| 1327 |
},
|
| 1328 |
{
|
| 1329 |
+
"epoch": 1.4901992548193748,
|
| 1330 |
+
"grad_norm": 2.717992067337036,
|
| 1331 |
+
"learning_rate": 8.933333333333334e-07,
|
| 1332 |
+
"loss": 0.2779,
|
| 1333 |
"step": 4600
|
| 1334 |
},
|
| 1335 |
{
|
| 1336 |
+
"epoch": 1.4982990442248503,
|
| 1337 |
+
"grad_norm": 2.852125406265259,
|
| 1338 |
+
"learning_rate": 8.37777777777778e-07,
|
| 1339 |
+
"loss": 0.2596,
|
| 1340 |
"step": 4625
|
| 1341 |
},
|
| 1342 |
{
|
| 1343 |
+
"epoch": 1.5063988336303256,
|
| 1344 |
+
"grad_norm": 2.2071590423583984,
|
| 1345 |
+
"learning_rate": 7.822222222222223e-07,
|
| 1346 |
+
"loss": 0.2765,
|
| 1347 |
"step": 4650
|
| 1348 |
},
|
| 1349 |
{
|
| 1350 |
+
"epoch": 1.514498623035801,
|
| 1351 |
+
"grad_norm": 2.871159315109253,
|
| 1352 |
+
"learning_rate": 7.266666666666668e-07,
|
| 1353 |
+
"loss": 0.2594,
|
| 1354 |
"step": 4675
|
| 1355 |
},
|
| 1356 |
{
|
| 1357 |
+
"epoch": 1.5225984124412766,
|
| 1358 |
+
"grad_norm": 2.287871837615967,
|
| 1359 |
+
"learning_rate": 6.711111111111111e-07,
|
| 1360 |
+
"loss": 0.2703,
|
| 1361 |
"step": 4700
|
| 1362 |
},
|
| 1363 |
{
|
| 1364 |
+
"epoch": 1.5306982018467519,
|
| 1365 |
+
"grad_norm": 2.345743417739868,
|
| 1366 |
+
"learning_rate": 6.155555555555556e-07,
|
| 1367 |
+
"loss": 0.2641,
|
| 1368 |
"step": 4725
|
| 1369 |
},
|
| 1370 |
{
|
| 1371 |
+
"epoch": 1.5387979912522276,
|
| 1372 |
+
"grad_norm": 2.734402656555176,
|
| 1373 |
+
"learning_rate": 5.6e-07,
|
| 1374 |
+
"loss": 0.2644,
|
| 1375 |
"step": 4750
|
| 1376 |
},
|
| 1377 |
{
|
| 1378 |
+
"epoch": 1.546897780657703,
|
| 1379 |
+
"grad_norm": 2.6672370433807373,
|
| 1380 |
+
"learning_rate": 5.044444444444445e-07,
|
| 1381 |
+
"loss": 0.2568,
|
| 1382 |
"step": 4775
|
| 1383 |
},
|
| 1384 |
{
|
| 1385 |
+
"epoch": 1.5549975700631784,
|
| 1386 |
+
"grad_norm": 2.9469447135925293,
|
| 1387 |
+
"learning_rate": 4.488888888888889e-07,
|
| 1388 |
+
"loss": 0.2646,
|
| 1389 |
"step": 4800
|
| 1390 |
},
|
| 1391 |
{
|
| 1392 |
+
"epoch": 1.563097359468654,
|
| 1393 |
+
"grad_norm": 3.1269147396087646,
|
| 1394 |
+
"learning_rate": 3.9333333333333336e-07,
|
| 1395 |
+
"loss": 0.2643,
|
| 1396 |
"step": 4825
|
| 1397 |
},
|
| 1398 |
{
|
| 1399 |
+
"epoch": 1.5711971488741292,
|
| 1400 |
+
"grad_norm": 2.543921709060669,
|
| 1401 |
+
"learning_rate": 3.3777777777777777e-07,
|
| 1402 |
+
"loss": 0.2738,
|
| 1403 |
"step": 4850
|
| 1404 |
},
|
| 1405 |
{
|
| 1406 |
+
"epoch": 1.5792969382796047,
|
| 1407 |
+
"grad_norm": 2.7174246311187744,
|
| 1408 |
+
"learning_rate": 2.822222222222222e-07,
|
| 1409 |
+
"loss": 0.2439,
|
| 1410 |
"step": 4875
|
| 1411 |
},
|
| 1412 |
{
|
| 1413 |
+
"epoch": 1.5873967276850802,
|
| 1414 |
+
"grad_norm": 2.381450891494751,
|
| 1415 |
+
"learning_rate": 2.266666666666667e-07,
|
| 1416 |
+
"loss": 0.2773,
|
| 1417 |
"step": 4900
|
| 1418 |
},
|
| 1419 |
{
|
| 1420 |
+
"epoch": 1.5954965170905555,
|
| 1421 |
+
"grad_norm": 2.9978244304656982,
|
| 1422 |
+
"learning_rate": 1.7111111111111114e-07,
|
| 1423 |
+
"loss": 0.2653,
|
| 1424 |
"step": 4925
|
| 1425 |
},
|
| 1426 |
{
|
| 1427 |
+
"epoch": 1.6035963064960312,
|
| 1428 |
+
"grad_norm": 2.594560384750366,
|
| 1429 |
+
"learning_rate": 1.1555555555555556e-07,
|
| 1430 |
+
"loss": 0.2472,
|
| 1431 |
"step": 4950
|
| 1432 |
},
|
| 1433 |
{
|
| 1434 |
+
"epoch": 1.6116960959015065,
|
| 1435 |
+
"grad_norm": 2.8399264812469482,
|
| 1436 |
+
"learning_rate": 6.000000000000001e-08,
|
| 1437 |
+
"loss": 0.258,
|
| 1438 |
"step": 4975
|
| 1439 |
},
|
| 1440 |
{
|
| 1441 |
+
"epoch": 1.619795885306982,
|
| 1442 |
+
"grad_norm": 2.9382212162017822,
|
| 1443 |
+
"learning_rate": 4.444444444444445e-09,
|
| 1444 |
+
"loss": 0.274,
|
| 1445 |
"step": 5000
|
| 1446 |
},
|
| 1447 |
{
|
| 1448 |
+
"epoch": 1.619795885306982,
|
| 1449 |
+
"eval_loss": 0.3738669753074646,
|
| 1450 |
+
"eval_runtime": 1856.8663,
|
| 1451 |
+
"eval_samples_per_second": 2.099,
|
| 1452 |
+
"eval_steps_per_second": 0.131,
|
| 1453 |
+
"eval_wer": 0.29154381412902464,
|
| 1454 |
"step": 5000
|
| 1455 |
},
|
| 1456 |
{
|
| 1457 |
+
"epoch": 1.619795885306982,
|
| 1458 |
"step": 5000,
|
| 1459 |
+
"total_flos": 5.435419715783885e+20,
|
| 1460 |
+
"train_loss": 0.4225531764030456,
|
| 1461 |
+
"train_runtime": 64153.0147,
|
| 1462 |
+
"train_samples_per_second": 2.494,
|
| 1463 |
+
"train_steps_per_second": 0.078
|
| 1464 |
}
|
| 1465 |
],
|
| 1466 |
"logging_steps": 25,
|
| 1467 |
"max_steps": 5000,
|
| 1468 |
"num_input_tokens_seen": 0,
|
| 1469 |
+
"num_train_epochs": 2,
|
| 1470 |
"save_steps": 1000,
|
| 1471 |
"stateful_callbacks": {
|
| 1472 |
"TrainerControl": {
|
|
|
|
| 1480 |
"attributes": {}
|
| 1481 |
}
|
| 1482 |
},
|
| 1483 |
+
"total_flos": 5.435419715783885e+20,
|
| 1484 |
"train_batch_size": 16,
|
| 1485 |
"trial_name": null,
|
| 1486 |
"trial_params": null
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d3ec45a0f94508361d7a8afc788e43452d0d48a3e9839538d81547b065babf44
|
| 3 |
+
size 5560
|