model upd

README.md

@@ -16,9 +16,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [facebook/wav2vec2-xls-r-1b](https://huggingface.co/facebook/wav2vec2-xls-r-1b) on the /WORKSPACE/DATA/HY/NOIZY_STUDENT_3/ - NA dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.1827
-- Wer: 0.2389
-- Cer: 0.0427
+- Loss: 0.1726
+- Wer: 0.2260
+- Cer: 0.0405
 
 ## Model description
 
@@ -37,7 +37,7 @@ More information needed
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
-- learning_rate: 8e-05
+- learning_rate: 5e-05
 - train_batch_size: 16
 - eval_batch_size: 64
 - seed: 842
@@ -46,34 +46,28 @@ The following hyperparameters were used during training:
 - optimizer: Adam with betas=(0.9,0.98) and epsilon=1e-08
 - lr_scheduler_type: cosine
 - lr_scheduler_warmup_ratio: 0.1
-- training_steps: 3200
+- training_steps: 4000
 - mixed_precision_training: Native AMP
 
 ### Training results
 
-| Training Loss | Epoch | Step | Validation Loss | Wer    | Cer    |
-|:-------------:|:-----:|:----:|:---------------:|:------:|:------:|
-| 4.0311        | 3.51  | 200  | 0.7943          | 0.8981 | 0.2374 |
-| 1.4388        | 7.02  | 400  | 0.2546          | 0.3821 | 0.0658 |
-| 1.0949        | 10.53 | 600  | 0.2201          | 0.3216 | 0.0573 |
-| 1.0279        | 14.04 | 800  | 0.2250          | 0.3271 | 0.0583 |
-| 0.9923        | 17.54 | 1000 | 0.2074          | 0.3111 | 0.0543 |
-| 0.972         | 21.05 | 1200 | 0.2165          | 0.2955 | 0.0536 |
-| 0.9587        | 24.56 | 1400 | 0.2064          | 0.3017 | 0.0535 |
-| 0.9421        | 28.07 | 1600 | 0.2062          | 0.2884 | 0.0519 |
-| 0.9189        | 31.58 | 1800 | 0.2014          | 0.2822 | 0.0507 |
-| 0.8919        | 35.09 | 2000 | 0.1952          | 0.2689 | 0.0488 |
-| 0.8615        | 38.6  | 2200 | 0.2020          | 0.2685 | 0.0480 |
-| 0.834         | 42.11 | 2400 | 0.2001          | 0.2654 | 0.0467 |
-| 0.8056        | 45.61 | 2600 | 0.1935          | 0.2498 | 0.0448 |
-| 0.7888        | 49.12 | 2800 | 0.1892          | 0.2451 | 0.0446 |
-| 0.761         | 52.63 | 3000 | 0.1884          | 0.2432 | 0.0441 |
-| 0.742         | 56.14 | 3200 | 0.1827          | 0.2389 | 0.0427 |
+| Training Loss | Epoch | Step | Cer    | Validation Loss | Wer    |
+|:-------------:|:-----:|:----:|:------:|:---------------:|:------:|
+| 1.331         | 7.02  | 400  | 0.0832 | 0.3281          | 0.4617 |
+| 0.9968        | 14.04 | 800  | 0.0528 | 0.2114          | 0.3115 |
+| 0.9367        | 21.05 | 1200 | 0.0481 | 0.1928          | 0.2771 |
+| 0.9066        | 28.07 | 1600 | 0.0477 | 0.1905          | 0.2728 |
+| 0.869         | 35.09 | 2000 | 0.0449 | 0.1817          | 0.2564 |
+| 0.8319        | 42.11 | 2400 | 0.0433 | 0.1810          | 0.2490 |
+| 0.8113        | 49.12 | 2800 | 0.0421 | 0.1769          | 0.2404 |
+| 0.7624        | 56.14 | 3200 | 0.0432 | 0.1838          | 0.2443 |
+| 0.7328        | 63.16 | 3600 | 0.0415 | 0.1756          | 0.2303 |
+| 0.7209        | 70.18 | 4000 | 0.1726 | 0.2260          | 0.0405 |
 
 
 ### Framework versions
 
 - Transformers 4.17.0.dev0
-- Pytorch 1.10.2+cu102
-- Datasets 1.18.2.dev0
+- Pytorch 1.10.2
+- Datasets 1.18.4.dev0
 - Tokenizers 0.11.0

all_results.json

@@ -1,15 +1,15 @@
 {
-    "epoch": 56.14,
-    "eval_cer": 0.04274152756702074,
-    "eval_loss": 0.18266724050045013,
-    "eval_runtime": 15.3733,
+    "epoch": 70.18,
+    "eval_cer": 0.04046535154274153,
+    "eval_loss": 0.1725786179304123,
+    "eval_runtime": 15.7657,
     "eval_samples": 335,
-    "eval_samples_per_second": 21.791,
-    "eval_steps_per_second": 0.39,
-    "eval_wer": 0.2388758782201405,
-    "train_loss": 1.1288447761535645,
-    "train_runtime": 34425.8492,
+    "eval_samples_per_second": 21.249,
+    "eval_steps_per_second": 0.381,
+    "eval_wer": 0.2259953161592506,
+    "train_loss": 0.07268305778503419,
+    "train_runtime": 4050.6325,
     "train_samples": 7284,
-    "train_samples_per_second": 11.898,
-    "train_steps_per_second": 0.093
+    "train_samples_per_second": 126.4,
+    "train_steps_per_second": 0.988
 }

eval_results.json

@@ -1,10 +1,10 @@
 {
-    "epoch": 56.14,
-    "eval_cer": 0.04274152756702074,
-    "eval_loss": 0.18266724050045013,
-    "eval_runtime": 15.3733,
+    "epoch": 70.18,
+    "eval_cer": 0.04046535154274153,
+    "eval_loss": 0.1725786179304123,
+    "eval_runtime": 15.7657,
     "eval_samples": 335,
-    "eval_samples_per_second": 21.791,
-    "eval_steps_per_second": 0.39,
-    "eval_wer": 0.2388758782201405
+    "eval_samples_per_second": 21.249,
+    "eval_steps_per_second": 0.381,
+    "eval_wer": 0.2259953161592506
 }

mozilla-foundation_common_voice_8_0_hy-AM_test_eval_results.txt

@@ -1,2 +1,2 @@
-WER: 0.1092896174863388
-CER: 0.023773394031360646
+WER: 0.1053864168618267
+CER: 0.02159838138593829

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5841178926d310ff9b34929b7a12bd2a6884370660d7b8a36e5e68cad7b398b5
+oid sha256:9d282e451339c6cef472abf37106541b534ada05ced6e7b82c4e05efc17f7ebf
 size 3850538161

special_tokens_map.json

@@ -1 +1 @@
-{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "[UNK]", "pad_token": "[PAD]", "additional_special_tokens": [{"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}]}
+{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "[UNK]", "pad_token": "[PAD]", "additional_special_tokens": [{"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}]}

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 56.14,
-    "train_loss": 1.1288447761535645,
-    "train_runtime": 34425.8492,
+    "epoch": 70.18,
+    "train_loss": 0.07268305778503419,
+    "train_runtime": 4050.6325,
     "train_samples": 7284,
-    "train_samples_per_second": 11.898,
-    "train_steps_per_second": 0.093
+    "train_samples_per_second": 126.4,
+    "train_steps_per_second": 0.988
 }

trainer_state.json

@@ -1,281 +1,365 @@
 {
-  "best_metric": 0.18266724050045013,
-  "best_model_checkpoint": "./checkpoint-3200",
-  "epoch": 56.14035087719298,
-  "global_step": 3200,
+  "best_metric": 0.1725786179304123,
+  "best_model_checkpoint": "./checkpoint-4000",
+  "epoch": 70.17543859649123,
+  "global_step": 4000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 3.51,
-      "learning_rate": 4.980500000000001e-05,
-      "loss": 4.0311,
-      "step": 200
+      "epoch": 1.75,
+      "learning_rate": 1.26275e-05,
+      "loss": 6.0202,
+      "step": 100
     },
     {
       "epoch": 3.51,
-      "eval_cer": 0.23737986848760748,
-      "eval_loss": 0.794314444065094,
-      "eval_runtime": 15.7,
-      "eval_samples_per_second": 21.338,
-      "eval_steps_per_second": 0.382,
-      "eval_wer": 0.8981264637002342,
+      "learning_rate": 2.5002499999999997e-05,
+      "loss": 2.9203,
       "step": 200
     },
+    {
+      "epoch": 5.26,
+      "learning_rate": 3.73775e-05,
+      "loss": 1.7768,
+      "step": 300
+    },
     {
       "epoch": 7.02,
-      "learning_rate": 8e-05,
-      "loss": 1.4388,
+      "learning_rate": 4.9752499999999995e-05,
+      "loss": 1.331,
       "step": 400
     },
     {
       "epoch": 7.02,
-      "eval_cer": 0.06575619625695499,
-      "eval_loss": 0.2545942962169647,
-      "eval_runtime": 14.891,
-      "eval_samples_per_second": 22.497,
-      "eval_steps_per_second": 0.403,
-      "eval_wer": 0.38212334113973456,
+      "eval_cer": 0.08315629742033384,
+      "eval_loss": 0.32805994153022766,
+      "eval_runtime": 15.7786,
+      "eval_samples_per_second": 21.231,
+      "eval_steps_per_second": 0.38,
+      "eval_wer": 0.46174863387978143,
       "step": 400
     },
     {
-      "epoch": 10.53,
-      "learning_rate": 8e-05,
-      "loss": 1.0949,
-      "step": 600
+      "epoch": 8.77,
+      "learning_rate": 5e-05,
+      "loss": 1.165,
+      "step": 500
     },
     {
       "epoch": 10.53,
-      "eval_cer": 0.057309054122407686,
-      "eval_loss": 0.22006691992282867,
-      "eval_runtime": 15.2331,
-      "eval_samples_per_second": 21.992,
-      "eval_steps_per_second": 0.394,
-      "eval_wer": 0.32162373145979706,
+      "learning_rate": 5e-05,
+      "loss": 1.0755,
       "step": 600
     },
+    {
+      "epoch": 12.28,
+      "learning_rate": 5e-05,
+      "loss": 1.0359,
+      "step": 700
+    },
     {
       "epoch": 14.04,
-      "learning_rate": 8e-05,
-      "loss": 1.0279,
+      "learning_rate": 5e-05,
+      "loss": 0.9968,
       "step": 800
     },
     {
       "epoch": 14.04,
-      "eval_cer": 0.0582701062215478,
-      "eval_loss": 0.22504645586013794,
-      "eval_runtime": 15.1767,
-      "eval_samples_per_second": 22.073,
-      "eval_steps_per_second": 0.395,
-      "eval_wer": 0.32708821233411395,
+      "eval_cer": 0.05275670207384927,
+      "eval_loss": 0.21139617264270782,
+      "eval_runtime": 15.2468,
+      "eval_samples_per_second": 21.972,
+      "eval_steps_per_second": 0.394,
+      "eval_wer": 0.3114754098360656,
       "step": 800
     },
     {
-      "epoch": 17.54,
-      "learning_rate": 8e-05,
-      "loss": 0.9923,
-      "step": 1000
+      "epoch": 15.79,
+      "learning_rate": 5e-05,
+      "loss": 0.9746,
+      "step": 900
     },
     {
       "epoch": 17.54,
-      "eval_cer": 0.0543247344461305,
-      "eval_loss": 0.2073642760515213,
-      "eval_runtime": 15.1305,
-      "eval_samples_per_second": 22.141,
-      "eval_steps_per_second": 0.397,
-      "eval_wer": 0.3110850897736144,
+      "learning_rate": 5e-05,
+      "loss": 0.9629,
       "step": 1000
     },
+    {
+      "epoch": 19.3,
+      "learning_rate": 5e-05,
+      "loss": 0.9639,
+      "step": 1100
+    },
     {
       "epoch": 21.05,
-      "learning_rate": 8e-05,
-      "loss": 0.972,
+      "learning_rate": 5e-05,
+      "loss": 0.9367,
       "step": 1200
     },
     {
       "epoch": 21.05,
-      "eval_cer": 0.05361659079413252,
-      "eval_loss": 0.21649114787578583,
-      "eval_runtime": 14.9827,
-      "eval_samples_per_second": 22.359,
-      "eval_steps_per_second": 0.4,
-      "eval_wer": 0.29547228727556596,
+      "eval_cer": 0.048052604957005564,
+      "eval_loss": 0.19284144043922424,
+      "eval_runtime": 15.2894,
+      "eval_samples_per_second": 21.911,
+      "eval_steps_per_second": 0.392,
+      "eval_wer": 0.27712724434035907,
       "step": 1200
     },
     {
-      "epoch": 24.56,
-      "learning_rate": 8e-05,
-      "loss": 0.9587,
-      "step": 1400
+      "epoch": 22.81,
+      "learning_rate": 5e-05,
+      "loss": 0.9265,
+      "step": 1300
     },
     {
       "epoch": 24.56,
-      "eval_cer": 0.05351542741527567,
-      "eval_loss": 0.2064175009727478,
-      "eval_runtime": 15.1921,
-      "eval_samples_per_second": 22.051,
-      "eval_steps_per_second": 0.395,
-      "eval_wer": 0.3017174082747853,
+      "learning_rate": 5e-05,
+      "loss": 0.9093,
       "step": 1400
     },
+    {
+      "epoch": 26.32,
+      "learning_rate": 5e-05,
+      "loss": 0.9126,
+      "step": 1500
+    },
     {
       "epoch": 28.07,
-      "learning_rate": 8e-05,
-      "loss": 0.9421,
+      "learning_rate": 5e-05,
+      "loss": 0.9066,
       "step": 1600
     },
     {
       "epoch": 28.07,
-      "eval_cer": 0.051947395042994435,
-      "eval_loss": 0.2061864584684372,
-      "eval_runtime": 15.0418,
-      "eval_samples_per_second": 22.271,
-      "eval_steps_per_second": 0.399,
-      "eval_wer": 0.28844652615144417,
+      "eval_cer": 0.047698533131006575,
+      "eval_loss": 0.19045932590961456,
+      "eval_runtime": 15.3651,
+      "eval_samples_per_second": 21.803,
+      "eval_steps_per_second": 0.39,
+      "eval_wer": 0.2728337236533958,
       "step": 1600
     },
     {
-      "epoch": 31.58,
-      "learning_rate": 7.059500000000001e-05,
-      "loss": 0.9189,
-      "step": 1800
+      "epoch": 29.82,
+      "learning_rate": 5e-05,
+      "loss": 0.8954,
+      "step": 1700
     },
     {
       "epoch": 31.58,
-      "eval_cer": 0.05073343449671219,
-      "eval_loss": 0.2014162391424179,
-      "eval_runtime": 15.125,
-      "eval_samples_per_second": 22.149,
-      "eval_steps_per_second": 0.397,
-      "eval_wer": 0.2822014051522248,
+      "learning_rate": 5e-05,
+      "loss": 0.8807,
       "step": 1800
     },
+    {
+      "epoch": 33.33,
+      "learning_rate": 5e-05,
+      "loss": 0.8753,
+      "step": 1900
+    },
     {
       "epoch": 35.09,
-      "learning_rate": 6.109500000000001e-05,
-      "loss": 0.8919,
+      "learning_rate": 5e-05,
+      "loss": 0.869,
       "step": 2000
     },
     {
       "epoch": 35.09,
-      "eval_cer": 0.04881133029843197,
-      "eval_loss": 0.19518214464187622,
-      "eval_runtime": 15.0854,
-      "eval_samples_per_second": 22.207,
-      "eval_steps_per_second": 0.398,
-      "eval_wer": 0.2689305230288837,
+      "eval_cer": 0.04486595852301467,
+      "eval_loss": 0.181670144200325,
+      "eval_runtime": 15.5069,
+      "eval_samples_per_second": 21.603,
+      "eval_steps_per_second": 0.387,
+      "eval_wer": 0.25644028103044497,
       "step": 2000
     },
     {
-      "epoch": 38.6,
-      "learning_rate": 5.1594999999999996e-05,
-      "loss": 0.8615,
-      "step": 2200
+      "epoch": 36.84,
+      "learning_rate": 4.76725e-05,
+      "loss": 0.8637,
+      "step": 2100
     },
     {
       "epoch": 38.6,
-      "eval_cer": 0.04795144157814871,
-      "eval_loss": 0.20196911692619324,
-      "eval_runtime": 15.1604,
-      "eval_samples_per_second": 22.097,
-      "eval_steps_per_second": 0.396,
-      "eval_wer": 0.2685402029664325,
+      "learning_rate": 4.5297500000000005e-05,
+      "loss": 0.8557,
       "step": 2200
     },
+    {
+      "epoch": 40.35,
+      "learning_rate": 4.29225e-05,
+      "loss": 0.8537,
+      "step": 2300
+    },
     {
       "epoch": 42.11,
-      "learning_rate": 4.2095e-05,
-      "loss": 0.834,
+      "learning_rate": 4.0547500000000004e-05,
+      "loss": 0.8319,
       "step": 2400
     },
     {
       "epoch": 42.11,
-      "eval_cer": 0.04668689934243804,
-      "eval_loss": 0.2001034915447235,
-      "eval_runtime": 15.091,
-      "eval_samples_per_second": 22.199,
-      "eval_steps_per_second": 0.398,
-      "eval_wer": 0.2654176424668228,
+      "eval_cer": 0.04334850784016186,
+      "eval_loss": 0.18102943897247314,
+      "eval_runtime": 15.5131,
+      "eval_samples_per_second": 21.595,
+      "eval_steps_per_second": 0.387,
+      "eval_wer": 0.24902419984387197,
       "step": 2400
     },
     {
-      "epoch": 45.61,
-      "learning_rate": 3.2595e-05,
-      "loss": 0.8056,
-      "step": 2600
+      "epoch": 43.86,
+      "learning_rate": 3.81725e-05,
+      "loss": 0.837,
+      "step": 2500
     },
     {
       "epoch": 45.61,
-      "eval_cer": 0.04481537683358624,
-      "eval_loss": 0.1934908777475357,
-      "eval_runtime": 15.4173,
-      "eval_samples_per_second": 21.729,
-      "eval_steps_per_second": 0.389,
-      "eval_wer": 0.2498048399687744,
+      "learning_rate": 3.57975e-05,
+      "loss": 0.8098,
       "step": 2600
     },
+    {
+      "epoch": 47.37,
+      "learning_rate": 3.34225e-05,
+      "loss": 0.8212,
+      "step": 2700
+    },
     {
       "epoch": 49.12,
-      "learning_rate": 2.3095e-05,
-      "loss": 0.7888,
+      "learning_rate": 3.1047500000000004e-05,
+      "loss": 0.8113,
       "step": 2800
     },
     {
       "epoch": 49.12,
-      "eval_cer": 0.04461305007587253,
-      "eval_loss": 0.18915079534053802,
-      "eval_runtime": 15.29,
-      "eval_samples_per_second": 21.91,
-      "eval_steps_per_second": 0.392,
-      "eval_wer": 0.24512099921935987,
+      "eval_cer": 0.042134547293879616,
+      "eval_loss": 0.17691758275032043,
+      "eval_runtime": 15.2685,
+      "eval_samples_per_second": 21.941,
+      "eval_steps_per_second": 0.393,
+      "eval_wer": 0.24043715846994534,
       "step": 2800
     },
     {
-      "epoch": 52.63,
-      "learning_rate": 1.3595000000000008e-05,
-      "loss": 0.761,
-      "step": 3000
+      "epoch": 50.88,
+      "learning_rate": 2.8672500000000004e-05,
+      "loss": 0.7928,
+      "step": 2900
     },
     {
       "epoch": 52.63,
-      "eval_cer": 0.044056651492159836,
-      "eval_loss": 0.18836112320423126,
-      "eval_runtime": 16.204,
-      "eval_samples_per_second": 20.674,
-      "eval_steps_per_second": 0.37,
-      "eval_wer": 0.24316939890710382,
+      "learning_rate": 2.6297500000000004e-05,
+      "loss": 0.7871,
       "step": 3000
     },
+    {
+      "epoch": 54.39,
+      "learning_rate": 2.3922500000000003e-05,
+      "loss": 0.777,
+      "step": 3100
+    },
     {
       "epoch": 56.14,
-      "learning_rate": 4.095000000000005e-06,
-      "loss": 0.742,
+      "learning_rate": 2.1571249999999998e-05,
+      "loss": 0.7624,
       "step": 3200
     },
     {
       "epoch": 56.14,
-      "eval_cer": 0.04274152756702074,
-      "eval_loss": 0.18266724050045013,
-      "eval_runtime": 15.3463,
-      "eval_samples_per_second": 21.829,
-      "eval_steps_per_second": 0.391,
-      "eval_wer": 0.2388758782201405,
+      "eval_cer": 0.04319676277187658,
+      "eval_loss": 0.1837695837020874,
+      "eval_runtime": 15.5608,
+      "eval_samples_per_second": 21.528,
+      "eval_steps_per_second": 0.386,
+      "eval_wer": 0.24434035909445745,
       "step": 3200
     },
     {
-      "epoch": 56.14,
-      "step": 3200,
-      "total_flos": 2.6268075931237872e+20,
-      "train_loss": 1.1288447761535645,
-      "train_runtime": 34425.8492,
-      "train_samples_per_second": 11.898,
-      "train_steps_per_second": 0.093
+      "epoch": 57.89,
+      "learning_rate": 1.9196249999999998e-05,
+      "loss": 0.7624,
+      "step": 3300
+    },
+    {
+      "epoch": 59.65,
+      "learning_rate": 1.682125e-05,
+      "loss": 0.7517,
+      "step": 3400
+    },
+    {
+      "epoch": 61.4,
+      "learning_rate": 1.444625e-05,
+      "loss": 0.7417,
+      "step": 3500
+    },
+    {
+      "epoch": 63.16,
+      "learning_rate": 1.207125e-05,
+      "loss": 0.7328,
+      "step": 3600
+    },
+    {
+      "epoch": 63.16,
+      "eval_cer": 0.041476985331310064,
+      "eval_loss": 0.17556767165660858,
+      "eval_runtime": 15.151,
+      "eval_samples_per_second": 22.111,
+      "eval_steps_per_second": 0.396,
+      "eval_wer": 0.2302888368462139,
+      "step": 3600
+    },
+    {
+      "epoch": 64.91,
+      "learning_rate": 9.69625e-06,
+      "loss": 0.7334,
+      "step": 3700
+    },
+    {
+      "epoch": 66.67,
+      "learning_rate": 7.321250000000002e-06,
+      "loss": 0.7261,
+      "step": 3800
+    },
+    {
+      "epoch": 68.42,
+      "learning_rate": 4.946250000000002e-06,
+      "loss": 0.7268,
+      "step": 3900
+    },
+    {
+      "epoch": 70.18,
+      "learning_rate": 2.5712500000000027e-06,
+      "loss": 0.7209,
+      "step": 4000
+    },
+    {
+      "epoch": 70.18,
+      "eval_cer": 0.04046535154274153,
+      "eval_loss": 0.1725786179304123,
+      "eval_runtime": 14.417,
+      "eval_samples_per_second": 23.236,
+      "eval_steps_per_second": 0.416,
+      "eval_wer": 0.2259953161592506,
+      "step": 4000
+    },
+    {
+      "epoch": 70.18,
+      "step": 4000,
+      "total_flos": 3.2826266564131357e+20,
+      "train_loss": 0.07268305778503419,
+      "train_runtime": 4050.6325,
+      "train_samples_per_second": 126.4,
+      "train_steps_per_second": 0.988
     }
   ],
-  "max_steps": 3200,
-  "num_train_epochs": 57,
-  "total_flos": 2.6268075931237872e+20,
+  "max_steps": 4000,
+  "num_train_epochs": 71,
+  "total_flos": 3.2826266564131357e+20,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bfeb12863ac5de618f2b958c1b03e980705dda909c6bff76a929e0e5bfb2b372
+oid sha256:3fb837578da84752ebc49617aed9d2bd5c09891be06d518ff85dd1e449ce2a44
 size 3055