Training complete - Strict Armenian WER: 42%

README.md

@@ -0,0 +1,58 @@
+---
+library_name: transformers
+language:
+- hy
+tags:
+- asr
+- audio
+- speech
+- whisper
+- low-resource
+- generated_from_trainer
+datasets:
+- Chillarmo/common_voice_20_armenian
+- mozilla-foundation/common_voice_20_0
+model-index:
+- name: checkpoint_9000
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# checkpoint_9000
+
+This model was trained from scratch on the Common Voice 20.0 dataset.
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 5e-05
+- train_batch_size: 8
+- eval_batch_size: 16
+- seed: 42
+- optimizer: Use adamw_torch_fused with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- lr_scheduler_type: linear
+- num_epochs: 3.0
+- mixed_precision_training: Native AMP
+
+### Framework versions
+
+- Transformers 4.56.2
+- Pytorch 2.8.0+cu129
+- Datasets 3.5.0
+- Tokenizers 0.22.1

config.json

@@ -0,0 +1,47 @@
+{
+  "activation_dropout": 0.0,
+  "activation_function": "gelu",
+  "apply_spec_augment": false,
+  "architectures": [
+    "WhisperForConditionalGeneration"
+  ],
+  "attention_dropout": 0.0,
+  "begin_suppress_tokens": null,
+  "bos_token_id": 1,
+  "classifier_proj_size": 256,
+  "d_model": 384,
+  "decoder_attention_heads": 6,
+  "decoder_ffn_dim": 1024,
+  "decoder_layerdrop": 0.0,
+  "decoder_layers": 2,
+  "decoder_start_token_id": 1,
+  "dropout": 0.0,
+  "dtype": "float16",
+  "encoder_attention_heads": 6,
+  "encoder_ffn_dim": 1536,
+  "encoder_layerdrop": 0.0,
+  "encoder_layers": 4,
+  "eos_token_id": 2,
+  "forced_decoder_ids": null,
+  "init_std": 0.02,
+  "is_encoder_decoder": true,
+  "mask_feature_length": 10,
+  "mask_feature_min_masks": 0,
+  "mask_feature_prob": 0.0,
+  "mask_time_length": 10,
+  "mask_time_min_masks": 2,
+  "mask_time_prob": 0.05,
+  "max_length": null,
+  "max_source_positions": 1500,
+  "max_target_positions": 448,
+  "median_filter_width": 7,
+  "model_type": "whisper",
+  "num_hidden_layers": 4,
+  "num_mel_bins": 80,
+  "pad_token_id": 0,
+  "scale_embedding": false,
+  "transformers_version": "4.56.2",
+  "use_cache": true,
+  "use_weighted_layer_sum": false,
+  "vocab_size": 5000
+}

generation_config.json

@@ -0,0 +1,12 @@
+{
+  "begin_suppress_tokens": [],
+  "bos_token_id": 1,
+  "decoder_start_token_id": 1,
+  "eos_token_id": 2,
+  "forced_decoder_ids": null,
+  "max_length": 448,
+  "pad_token_id": 0,
+  "suppress_tokens": [],
+  "transformers_version": "4.56.2",
+  "vocab_size": 5000
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:26f33a614a6e03eb2afbe7277fd9511446ce241dce7f1a70e23756b7b0fe7cbd
+size 28504056

optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:522cb173c1edaed339d0d2a44d3d383f4981a84159179948b3cad975dccebc86
+size 48340299

rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5ac7846d433abea5c16b7eeb3755a7faf76de4592a9b9d134ab5685daa27b433
+size 14645

scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:56410013d5a6584db199c4b19fe4268f15b8b3cb219a0f37341404083055bc16
+size 1383

scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6b11393457f05d4fbe28bb082192a3c5fbb59c96c0a7126f55f48cbb3b145578
+size 1465

special_tokens_map.json

@@ -0,0 +1,24 @@
+{
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<pad>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer_config.json

@@ -0,0 +1,45 @@
+{
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "extra_special_tokens": {},
+  "model_max_length": 448,
+  "pad_token": "<pad>",
+  "tokenizer_class": "WhisperTokenizer",
+  "unk_token": "<unk>"
+}

trainer_state.json

@@ -0,0 +1,664 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 11.734028683181226,
+  "eval_steps": 500,
+  "global_step": 9000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.1303780964797914,
+      "grad_norm": 2.199315071105957,
+      "learning_rate": 1.9800000000000004e-05,
+      "loss": 8.1376,
+      "step": 100
+    },
+    {
+      "epoch": 0.2607561929595828,
+      "grad_norm": 1.6398614645004272,
+      "learning_rate": 3.9800000000000005e-05,
+      "loss": 7.2341,
+      "step": 200
+    },
+    {
+      "epoch": 0.39113428943937417,
+      "grad_norm": 1.6936060190200806,
+      "learning_rate": 5.9800000000000003e-05,
+      "loss": 6.5612,
+      "step": 300
+    },
+    {
+      "epoch": 0.5215123859191656,
+      "grad_norm": 2.0208520889282227,
+      "learning_rate": 7.98e-05,
+      "loss": 6.2796,
+      "step": 400
+    },
+    {
+      "epoch": 0.651890482398957,
+      "grad_norm": 2.6271910667419434,
+      "learning_rate": 9.98e-05,
+      "loss": 6.0939,
+      "step": 500
+    },
+    {
+      "epoch": 0.7822685788787483,
+      "grad_norm": 3.186422348022461,
+      "learning_rate": 9.913913043478262e-05,
+      "loss": 5.8614,
+      "step": 600
+    },
+    {
+      "epoch": 0.9126466753585397,
+      "grad_norm": 2.3998301029205322,
+      "learning_rate": 9.826956521739131e-05,
+      "loss": 5.6819,
+      "step": 700
+    },
+    {
+      "epoch": 1.0430247718383312,
+      "grad_norm": 2.232499837875366,
+      "learning_rate": 9.74e-05,
+      "loss": 5.568,
+      "step": 800
+    },
+    {
+      "epoch": 1.1734028683181226,
+      "grad_norm": 3.5190041065216064,
+      "learning_rate": 9.65304347826087e-05,
+      "loss": 5.4076,
+      "step": 900
+    },
+    {
+      "epoch": 1.303780964797914,
+      "grad_norm": 3.0163352489471436,
+      "learning_rate": 9.566086956521739e-05,
+      "loss": 5.3031,
+      "step": 1000
+    },
+    {
+      "epoch": 1.4341590612777053,
+      "grad_norm": 2.686789035797119,
+      "learning_rate": 9.47913043478261e-05,
+      "loss": 5.1789,
+      "step": 1100
+    },
+    {
+      "epoch": 1.5645371577574967,
+      "grad_norm": 2.6344940662384033,
+      "learning_rate": 9.392173913043478e-05,
+      "loss": 5.0844,
+      "step": 1200
+    },
+    {
+      "epoch": 1.694915254237288,
+      "grad_norm": 3.4910781383514404,
+      "learning_rate": 9.305217391304349e-05,
+      "loss": 4.9929,
+      "step": 1300
+    },
+    {
+      "epoch": 1.8252933507170797,
+      "grad_norm": 3.5532472133636475,
+      "learning_rate": 9.218260869565217e-05,
+      "loss": 4.8998,
+      "step": 1400
+    },
+    {
+      "epoch": 1.9556714471968708,
+      "grad_norm": 3.246469497680664,
+      "learning_rate": 9.131304347826088e-05,
+      "loss": 4.7656,
+      "step": 1500
+    },
+    {
+      "epoch": 2.0860495436766624,
+      "grad_norm": 3.618093729019165,
+      "learning_rate": 9.044347826086958e-05,
+      "loss": 4.5793,
+      "step": 1600
+    },
+    {
+      "epoch": 2.2164276401564535,
+      "grad_norm": 4.264451026916504,
+      "learning_rate": 8.957391304347826e-05,
+      "loss": 4.4459,
+      "step": 1700
+    },
+    {
+      "epoch": 2.346805736636245,
+      "grad_norm": 4.198596477508545,
+      "learning_rate": 8.870434782608697e-05,
+      "loss": 4.3207,
+      "step": 1800
+    },
+    {
+      "epoch": 2.4771838331160367,
+      "grad_norm": 4.508902549743652,
+      "learning_rate": 8.783478260869565e-05,
+      "loss": 4.235,
+      "step": 1900
+    },
+    {
+      "epoch": 2.607561929595828,
+      "grad_norm": 4.511523723602295,
+      "learning_rate": 8.696521739130436e-05,
+      "loss": 4.1349,
+      "step": 2000
+    },
+    {
+      "epoch": 2.737940026075619,
+      "grad_norm": 3.903628349304199,
+      "learning_rate": 8.609565217391304e-05,
+      "loss": 4.0358,
+      "step": 2100
+    },
+    {
+      "epoch": 2.8683181225554106,
+      "grad_norm": 4.530259609222412,
+      "learning_rate": 8.523478260869565e-05,
+      "loss": 3.9168,
+      "step": 2200
+    },
+    {
+      "epoch": 2.9986962190352022,
+      "grad_norm": 4.071807861328125,
+      "learning_rate": 8.436521739130435e-05,
+      "loss": 3.8121,
+      "step": 2300
+    },
+    {
+      "epoch": 3.1290743155149934,
+      "grad_norm": 4.118851661682129,
+      "learning_rate": 8.349565217391306e-05,
+      "loss": 3.6089,
+      "step": 2400
+    },
+    {
+      "epoch": 3.259452411994785,
+      "grad_norm": 4.954625606536865,
+      "learning_rate": 8.262608695652174e-05,
+      "loss": 3.5162,
+      "step": 2500
+    },
+    {
+      "epoch": 3.389830508474576,
+      "grad_norm": 4.958930492401123,
+      "learning_rate": 8.175652173913044e-05,
+      "loss": 3.427,
+      "step": 2600
+    },
+    {
+      "epoch": 3.5202086049543677,
+      "grad_norm": 4.683836460113525,
+      "learning_rate": 8.088695652173913e-05,
+      "loss": 3.3498,
+      "step": 2700
+    },
+    {
+      "epoch": 3.6505867014341593,
+      "grad_norm": 5.2733259201049805,
+      "learning_rate": 8.001739130434783e-05,
+      "loss": 3.2661,
+      "step": 2800
+    },
+    {
+      "epoch": 3.7809647979139505,
+      "grad_norm": 5.038583278656006,
+      "learning_rate": 7.914782608695652e-05,
+      "loss": 3.1859,
+      "step": 2900
+    },
+    {
+      "epoch": 3.9113428943937416,
+      "grad_norm": 5.139358043670654,
+      "learning_rate": 7.827826086956522e-05,
+      "loss": 3.125,
+      "step": 3000
+    },
+    {
+      "epoch": 4.041720990873533,
+      "grad_norm": 5.076649188995361,
+      "learning_rate": 7.740869565217391e-05,
+      "loss": 2.9774,
+      "step": 3100
+    },
+    {
+      "epoch": 4.172099087353325,
+      "grad_norm": 7.2738165855407715,
+      "learning_rate": 7.653913043478261e-05,
+      "loss": 2.8133,
+      "step": 3200
+    },
+    {
+      "epoch": 4.302477183833116,
+      "grad_norm": 4.856014728546143,
+      "learning_rate": 7.56695652173913e-05,
+      "loss": 2.7578,
+      "step": 3300
+    },
+    {
+      "epoch": 4.432855280312907,
+      "grad_norm": 4.625192165374756,
+      "learning_rate": 7.48e-05,
+      "loss": 2.706,
+      "step": 3400
+    },
+    {
+      "epoch": 4.563233376792699,
+      "grad_norm": 5.898316860198975,
+      "learning_rate": 7.39304347826087e-05,
+      "loss": 2.6505,
+      "step": 3500
+    },
+    {
+      "epoch": 4.69361147327249,
+      "grad_norm": 5.6338324546813965,
+      "learning_rate": 7.30608695652174e-05,
+      "loss": 2.5716,
+      "step": 3600
+    },
+    {
+      "epoch": 4.823989569752282,
+      "grad_norm": 4.824554443359375,
+      "learning_rate": 7.219130434782609e-05,
+      "loss": 2.4845,
+      "step": 3700
+    },
+    {
+      "epoch": 4.9543676662320735,
+      "grad_norm": 6.3988037109375,
+      "learning_rate": 7.132173913043479e-05,
+      "loss": 2.4531,
+      "step": 3800
+    },
+    {
+      "epoch": 5.084745762711864,
+      "grad_norm": 4.477128982543945,
+      "learning_rate": 7.045217391304348e-05,
+      "loss": 2.3122,
+      "step": 3900
+    },
+    {
+      "epoch": 5.215123859191656,
+      "grad_norm": 5.06989049911499,
+      "learning_rate": 6.958260869565218e-05,
+      "loss": 2.2015,
+      "step": 4000
+    },
+    {
+      "epoch": 5.345501955671447,
+      "grad_norm": 4.576299667358398,
+      "learning_rate": 6.871304347826087e-05,
+      "loss": 2.1651,
+      "step": 4100
+    },
+    {
+      "epoch": 5.475880052151239,
+      "grad_norm": 4.288216590881348,
+      "learning_rate": 6.785217391304349e-05,
+      "loss": 2.1132,
+      "step": 4200
+    },
+    {
+      "epoch": 5.60625814863103,
+      "grad_norm": 6.912166118621826,
+      "learning_rate": 6.698260869565218e-05,
+      "loss": 2.0712,
+      "step": 4300
+    },
+    {
+      "epoch": 5.736636245110821,
+      "grad_norm": 5.21943998336792,
+      "learning_rate": 6.611304347826088e-05,
+      "loss": 2.0318,
+      "step": 4400
+    },
+    {
+      "epoch": 5.867014341590613,
+      "grad_norm": 6.814871311187744,
+      "learning_rate": 6.524347826086957e-05,
+      "loss": 1.996,
+      "step": 4500
+    },
+    {
+      "epoch": 5.9973924380704045,
+      "grad_norm": 6.987309455871582,
+      "learning_rate": 6.437391304347827e-05,
+      "loss": 1.9435,
+      "step": 4600
+    },
+    {
+      "epoch": 6.127770534550195,
+      "grad_norm": 6.552413463592529,
+      "learning_rate": 6.350434782608696e-05,
+      "loss": 1.7934,
+      "step": 4700
+    },
+    {
+      "epoch": 6.258148631029987,
+      "grad_norm": 4.565166473388672,
+      "learning_rate": 6.263478260869565e-05,
+      "loss": 1.7635,
+      "step": 4800
+    },
+    {
+      "epoch": 6.388526727509778,
+      "grad_norm": 6.433986186981201,
+      "learning_rate": 6.176521739130436e-05,
+      "loss": 1.7403,
+      "step": 4900
+    },
+    {
+      "epoch": 6.51890482398957,
+      "grad_norm": 4.71969747543335,
+      "learning_rate": 6.0895652173913044e-05,
+      "loss": 1.7153,
+      "step": 5000
+    },
+    {
+      "epoch": 6.6492829204693615,
+      "grad_norm": 5.415853023529053,
+      "learning_rate": 6.002608695652174e-05,
+      "loss": 1.6563,
+      "step": 5100
+    },
+    {
+      "epoch": 6.779661016949152,
+      "grad_norm": 4.742384910583496,
+      "learning_rate": 5.9156521739130436e-05,
+      "loss": 1.6431,
+      "step": 5200
+    },
+    {
+      "epoch": 6.910039113428944,
+      "grad_norm": 4.704698085784912,
+      "learning_rate": 5.828695652173913e-05,
+      "loss": 1.6071,
+      "step": 5300
+    },
+    {
+      "epoch": 7.040417209908735,
+      "grad_norm": 4.532726764678955,
+      "learning_rate": 5.7417391304347834e-05,
+      "loss": 1.561,
+      "step": 5400
+    },
+    {
+      "epoch": 7.170795306388527,
+      "grad_norm": 4.465181827545166,
+      "learning_rate": 5.654782608695652e-05,
+      "loss": 1.4572,
+      "step": 5500
+    },
+    {
+      "epoch": 7.301173402868318,
+      "grad_norm": 5.061060428619385,
+      "learning_rate": 5.568695652173913e-05,
+      "loss": 1.4313,
+      "step": 5600
+    },
+    {
+      "epoch": 7.431551499348109,
+      "grad_norm": 4.221780300140381,
+      "learning_rate": 5.481739130434783e-05,
+      "loss": 1.4099,
+      "step": 5700
+    },
+    {
+      "epoch": 7.561929595827901,
+      "grad_norm": 4.012993812561035,
+      "learning_rate": 5.394782608695652e-05,
+      "loss": 1.4044,
+      "step": 5800
+    },
+    {
+      "epoch": 7.6923076923076925,
+      "grad_norm": 5.935675621032715,
+      "learning_rate": 5.307826086956522e-05,
+      "loss": 1.3888,
+      "step": 5900
+    },
+    {
+      "epoch": 7.822685788787483,
+      "grad_norm": 5.962758541107178,
+      "learning_rate": 5.220869565217391e-05,
+      "loss": 1.3509,
+      "step": 6000
+    },
+    {
+      "epoch": 7.953063885267275,
+      "grad_norm": 5.295892238616943,
+      "learning_rate": 5.1339130434782615e-05,
+      "loss": 1.3376,
+      "step": 6100
+    },
+    {
+      "epoch": 8.083441981747066,
+      "grad_norm": 4.8437347412109375,
+      "learning_rate": 5.0469565217391304e-05,
+      "loss": 1.2682,
+      "step": 6200
+    },
+    {
+      "epoch": 8.213820078226858,
+      "grad_norm": 6.373332500457764,
+      "learning_rate": 4.96e-05,
+      "loss": 1.2201,
+      "step": 6300
+    },
+    {
+      "epoch": 8.34419817470665,
+      "grad_norm": 4.604928493499756,
+      "learning_rate": 4.8730434782608695e-05,
+      "loss": 1.2068,
+      "step": 6400
+    },
+    {
+      "epoch": 8.474576271186441,
+      "grad_norm": 6.433839797973633,
+      "learning_rate": 4.786086956521739e-05,
+      "loss": 1.197,
+      "step": 6500
+    },
+    {
+      "epoch": 8.604954367666233,
+      "grad_norm": 3.5552725791931152,
+      "learning_rate": 4.699130434782609e-05,
+      "loss": 1.1726,
+      "step": 6600
+    },
+    {
+      "epoch": 8.735332464146023,
+      "grad_norm": 8.83135986328125,
+      "learning_rate": 4.612173913043478e-05,
+      "loss": 1.1484,
+      "step": 6700
+    },
+    {
+      "epoch": 8.865710560625814,
+      "grad_norm": 4.828709602355957,
+      "learning_rate": 4.5252173913043485e-05,
+      "loss": 1.1583,
+      "step": 6800
+    },
+    {
+      "epoch": 8.996088657105606,
+      "grad_norm": 4.650803089141846,
+      "learning_rate": 4.438260869565218e-05,
+      "loss": 1.1481,
+      "step": 6900
+    },
+    {
+      "epoch": 9.126466753585397,
+      "grad_norm": 3.8928608894348145,
+      "learning_rate": 4.3513043478260876e-05,
+      "loss": 1.0397,
+      "step": 7000
+    },
+    {
+      "epoch": 9.256844850065189,
+      "grad_norm": 3.941554307937622,
+      "learning_rate": 4.2643478260869565e-05,
+      "loss": 1.0332,
+      "step": 7100
+    },
+    {
+      "epoch": 9.38722294654498,
+      "grad_norm": 5.031588077545166,
+      "learning_rate": 4.177391304347826e-05,
+      "loss": 1.0444,
+      "step": 7200
+    },
+    {
+      "epoch": 9.517601043024772,
+      "grad_norm": 5.682355880737305,
+      "learning_rate": 4.090434782608696e-05,
+      "loss": 1.0123,
+      "step": 7300
+    },
+    {
+      "epoch": 9.647979139504564,
+      "grad_norm": 4.108279705047607,
+      "learning_rate": 4.003478260869565e-05,
+      "loss": 1.0194,
+      "step": 7400
+    },
+    {
+      "epoch": 9.778357235984355,
+      "grad_norm": 3.3550074100494385,
+      "learning_rate": 3.916521739130435e-05,
+      "loss": 1.007,
+      "step": 7500
+    },
+    {
+      "epoch": 9.908735332464147,
+      "grad_norm": 4.076298713684082,
+      "learning_rate": 3.8295652173913044e-05,
+      "loss": 0.9973,
+      "step": 7600
+    },
+    {
+      "epoch": 10.039113428943937,
+      "grad_norm": 4.534595012664795,
+      "learning_rate": 3.742608695652174e-05,
+      "loss": 0.9646,
+      "step": 7700
+    },
+    {
+      "epoch": 10.169491525423728,
+      "grad_norm": 3.5908210277557373,
+      "learning_rate": 3.6556521739130435e-05,
+      "loss": 0.8984,
+      "step": 7800
+    },
+    {
+      "epoch": 10.29986962190352,
+      "grad_norm": 4.6488213539123535,
+      "learning_rate": 3.568695652173913e-05,
+      "loss": 0.8942,
+      "step": 7900
+    },
+    {
+      "epoch": 10.430247718383312,
+      "grad_norm": 4.129245758056641,
+      "learning_rate": 3.481739130434783e-05,
+      "loss": 0.8929,
+      "step": 8000
+    },
+    {
+      "epoch": 10.560625814863103,
+      "grad_norm": 3.645109176635742,
+      "learning_rate": 3.394782608695652e-05,
+      "loss": 0.8892,
+      "step": 8100
+    },
+    {
+      "epoch": 10.691003911342895,
+      "grad_norm": 4.7509565353393555,
+      "learning_rate": 3.307826086956522e-05,
+      "loss": 0.8937,
+      "step": 8200
+    },
+    {
+      "epoch": 10.821382007822686,
+      "grad_norm": 4.302902698516846,
+      "learning_rate": 3.2208695652173914e-05,
+      "loss": 0.8836,
+      "step": 8300
+    },
+    {
+      "epoch": 10.951760104302478,
+      "grad_norm": 4.266739845275879,
+      "learning_rate": 3.133913043478261e-05,
+      "loss": 0.8875,
+      "step": 8400
+    },
+    {
+      "epoch": 11.082138200782268,
+      "grad_norm": 3.6782920360565186,
+      "learning_rate": 3.0469565217391305e-05,
+      "loss": 0.8411,
+      "step": 8500
+    },
+    {
+      "epoch": 11.21251629726206,
+      "grad_norm": 3.9318060874938965,
+      "learning_rate": 2.96e-05,
+      "loss": 0.8063,
+      "step": 8600
+    },
+    {
+      "epoch": 11.342894393741851,
+      "grad_norm": 4.048793315887451,
+      "learning_rate": 2.8730434782608694e-05,
+      "loss": 0.7968,
+      "step": 8700
+    },
+    {
+      "epoch": 11.473272490221643,
+      "grad_norm": 4.4102888107299805,
+      "learning_rate": 2.786086956521739e-05,
+      "loss": 0.7983,
+      "step": 8800
+    },
+    {
+      "epoch": 11.603650586701434,
+      "grad_norm": 4.744139671325684,
+      "learning_rate": 2.6991304347826092e-05,
+      "loss": 0.7903,
+      "step": 8900
+    },
+    {
+      "epoch": 11.734028683181226,
+      "grad_norm": 4.147999286651611,
+      "learning_rate": 2.6121739130434788e-05,
+      "loss": 0.7997,
+      "step": 9000
+    }
+  ],
+  "logging_steps": 100,
+  "max_steps": 12000,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 16,
+  "save_steps": 3000,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 4.7976884011008e+18,
+  "train_batch_size": 32,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:50472477cc62d84b2d93e339afda70651261ba26b46389224cbac1258329a178
+size 5969