Upload folder using huggingface_hub

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:843cb11d766af36ac71491251c6a59e1c18acf437ae2a81f66e4f4c75fe0c1fd
+oid sha256:f242ee095f7638b7463b942830986f1a347b04482aa841b4677146bd81ef888d
 size 1629436964

optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3c37b72249ab670a79f95e7a8a28b4295bc0d2474b70ef75088b7f94377a21cd
+oid sha256:64f655fa0dd0081f2268b67b7028d6c401ddca176d518cebfa125e630d681ef2
 size 3259184731

rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c615480f9132ca44d1a64f61d0ad33b4cfee708b0079b7d90597306c265b6e39
+oid sha256:7b8b8f78f8ab97f155a53e04798b283255c70c7d783ca3734914d92eb7738c00
 size 14244

scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:20f8ea15f450d448adf5f088f5e69b322ba185360623de1fad0e8c5fa0932135
+oid sha256:eb84971a8233865d3879ef75d98a5c44792abce4fbe173916e56341b9753aae6
 size 1064

trainer_state.json

@@ -1,378 +1,251 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 8.0,
+  "epoch": 6.0,
   "eval_steps": 500,
-  "global_step": 4232,
+  "global_step": 3342,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.19,
-      "grad_norm": 6.645576477050781,
-      "learning_rate": 1.9810964083175805e-05,
-      "loss": 0.5588,
+      "epoch": 0.18,
+      "grad_norm": 6.054772853851318,
+      "learning_rate": 1.9820466786355476e-05,
+      "loss": 0.5721,
       "step": 100
     },
     {
-      "epoch": 0.38,
-      "grad_norm": 4.662398815155029,
-      "learning_rate": 1.962192816635161e-05,
-      "loss": 0.3737,
+      "epoch": 0.36,
+      "grad_norm": 9.810357093811035,
+      "learning_rate": 1.9640933572710953e-05,
+      "loss": 0.3846,
       "step": 200
     },
     {
-      "epoch": 0.57,
-      "grad_norm": 5.282985687255859,
-      "learning_rate": 1.9432892249527412e-05,
-      "loss": 0.2863,
+      "epoch": 0.54,
+      "grad_norm": 9.026122093200684,
+      "learning_rate": 1.9461400359066428e-05,
+      "loss": 0.3019,
       "step": 300
     },
     {
-      "epoch": 0.76,
-      "grad_norm": 1.5854848623275757,
-      "learning_rate": 1.9243856332703215e-05,
-      "loss": 0.323,
+      "epoch": 0.72,
+      "grad_norm": 11.95788288116455,
+      "learning_rate": 1.9281867145421905e-05,
+      "loss": 0.3071,
       "step": 400
     },
     {
-      "epoch": 0.95,
-      "grad_norm": 19.09914779663086,
-      "learning_rate": 1.905482041587902e-05,
-      "loss": 0.2956,
+      "epoch": 0.9,
+      "grad_norm": 15.329608917236328,
+      "learning_rate": 1.910233393177738e-05,
+      "loss": 0.2907,
       "step": 500
     },
     {
-      "epoch": 1.0,
-      "eval_loss": 0.20477035641670227,
-      "eval_runtime": 4.245,
-      "eval_samples_per_second": 91.638,
-      "eval_steps_per_second": 6.596,
-      "step": 529
-    },
-    {
-      "epoch": 1.13,
-      "grad_norm": 26.59105682373047,
-      "learning_rate": 1.8865784499054822e-05,
-      "loss": 0.2269,
+      "epoch": 1.08,
+      "grad_norm": 12.364314079284668,
+      "learning_rate": 1.8922800718132857e-05,
+      "loss": 0.2646,
       "step": 600
     },
     {
-      "epoch": 1.32,
-      "grad_norm": 9.394390106201172,
-      "learning_rate": 1.8676748582230626e-05,
-      "loss": 0.2393,
+      "epoch": 1.26,
+      "grad_norm": 14.555986404418945,
+      "learning_rate": 1.874326750448833e-05,
+      "loss": 0.2097,
       "step": 700
     },
     {
-      "epoch": 1.51,
-      "grad_norm": 13.779593467712402,
-      "learning_rate": 1.848771266540643e-05,
-      "loss": 0.2135,
+      "epoch": 1.44,
+      "grad_norm": 4.199421405792236,
+      "learning_rate": 1.8563734290843805e-05,
+      "loss": 0.2389,
       "step": 800
     },
     {
-      "epoch": 1.7,
-      "grad_norm": 0.6053704619407654,
-      "learning_rate": 1.8298676748582232e-05,
-      "loss": 0.2241,
+      "epoch": 1.62,
+      "grad_norm": 22.485984802246094,
+      "learning_rate": 1.8384201077199283e-05,
+      "loss": 0.2019,
       "step": 900
     },
     {
-      "epoch": 1.89,
-      "grad_norm": 16.190296173095703,
-      "learning_rate": 1.8109640831758036e-05,
-      "loss": 0.2355,
+      "epoch": 1.8,
+      "grad_norm": 9.688520431518555,
+      "learning_rate": 1.820466786355476e-05,
+      "loss": 0.2594,
       "step": 1000
     },
     {
-      "epoch": 2.0,
-      "eval_loss": 0.21921110153198242,
-      "eval_runtime": 4.2197,
-      "eval_samples_per_second": 92.187,
-      "eval_steps_per_second": 6.636,
-      "step": 1058
-    },
-    {
-      "epoch": 2.08,
-      "grad_norm": 0.10688259452581406,
-      "learning_rate": 1.792060491493384e-05,
-      "loss": 0.1909,
+      "epoch": 1.97,
+      "grad_norm": 18.052719116210938,
+      "learning_rate": 1.8025134649910235e-05,
+      "loss": 0.2044,
       "step": 1100
     },
     {
-      "epoch": 2.27,
-      "grad_norm": 29.609710693359375,
-      "learning_rate": 1.7731568998109643e-05,
-      "loss": 0.1637,
+      "epoch": 2.15,
+      "grad_norm": 0.7371789216995239,
+      "learning_rate": 1.7845601436265712e-05,
+      "loss": 0.1551,
       "step": 1200
     },
     {
-      "epoch": 2.46,
-      "grad_norm": 0.16163112223148346,
-      "learning_rate": 1.7542533081285446e-05,
-      "loss": 0.1828,
+      "epoch": 2.33,
+      "grad_norm": 20.938648223876953,
+      "learning_rate": 1.7666068222621186e-05,
+      "loss": 0.1463,
       "step": 1300
     },
     {
-      "epoch": 2.65,
-      "grad_norm": 2.127821207046509,
-      "learning_rate": 1.735349716446125e-05,
-      "loss": 0.1827,
+      "epoch": 2.51,
+      "grad_norm": 0.25227147340774536,
+      "learning_rate": 1.748653500897666e-05,
+      "loss": 0.1493,
       "step": 1400
     },
     {
-      "epoch": 2.84,
-      "grad_norm": 0.030160142108798027,
-      "learning_rate": 1.7164461247637053e-05,
-      "loss": 0.1528,
+      "epoch": 2.69,
+      "grad_norm": 0.27634137868881226,
+      "learning_rate": 1.7307001795332138e-05,
+      "loss": 0.1649,
       "step": 1500
     },
     {
-      "epoch": 3.0,
-      "eval_loss": 0.2590792179107666,
-      "eval_runtime": 4.2371,
-      "eval_samples_per_second": 91.808,
-      "eval_steps_per_second": 6.608,
-      "step": 1587
-    },
-    {
-      "epoch": 3.02,
-      "grad_norm": 0.0637928918004036,
-      "learning_rate": 1.6975425330812856e-05,
-      "loss": 0.1574,
+      "epoch": 2.87,
+      "grad_norm": 0.2588340938091278,
+      "learning_rate": 1.7127468581687616e-05,
+      "loss": 0.1521,
       "step": 1600
     },
     {
-      "epoch": 3.21,
-      "grad_norm": 0.01419603731483221,
-      "learning_rate": 1.678638941398866e-05,
-      "loss": 0.0918,
+      "epoch": 3.05,
+      "grad_norm": 0.05350634083151817,
+      "learning_rate": 1.694793536804309e-05,
+      "loss": 0.1343,
       "step": 1700
     },
     {
-      "epoch": 3.4,
-      "grad_norm": 9.31396770477295,
-      "learning_rate": 1.6597353497164463e-05,
-      "loss": 0.1175,
+      "epoch": 3.23,
+      "grad_norm": 0.02972230687737465,
+      "learning_rate": 1.6768402154398564e-05,
+      "loss": 0.1068,
       "step": 1800
     },
     {
-      "epoch": 3.59,
-      "grad_norm": 0.023506687954068184,
-      "learning_rate": 1.6408317580340267e-05,
-      "loss": 0.1094,
+      "epoch": 3.41,
+      "grad_norm": 0.09572970867156982,
+      "learning_rate": 1.658886894075404e-05,
+      "loss": 0.1151,
       "step": 1900
     },
     {
-      "epoch": 3.78,
-      "grad_norm": 13.4344482421875,
-      "learning_rate": 1.621928166351607e-05,
-      "loss": 0.1103,
+      "epoch": 3.59,
+      "grad_norm": 21.431325912475586,
+      "learning_rate": 1.6409335727109516e-05,
+      "loss": 0.1073,
       "step": 2000
     },
     {
-      "epoch": 3.97,
-      "grad_norm": 30.01926040649414,
-      "learning_rate": 1.6030245746691873e-05,
-      "loss": 0.1398,
+      "epoch": 3.77,
+      "grad_norm": 1.4688669443130493,
+      "learning_rate": 1.6229802513464993e-05,
+      "loss": 0.1098,
       "step": 2100
     },
     {
-      "epoch": 4.0,
-      "eval_loss": 0.33959057927131653,
-      "eval_runtime": 4.2467,
-      "eval_samples_per_second": 91.601,
-      "eval_steps_per_second": 6.593,
-      "step": 2116
-    },
-    {
-      "epoch": 4.16,
-      "grad_norm": 0.12888406217098236,
-      "learning_rate": 1.5841209829867677e-05,
-      "loss": 0.0536,
+      "epoch": 3.95,
+      "grad_norm": 19.461355209350586,
+      "learning_rate": 1.6050269299820467e-05,
+      "loss": 0.1238,
       "step": 2200
     },
     {
-      "epoch": 4.35,
-      "grad_norm": 0.06800971180200577,
-      "learning_rate": 1.565217391304348e-05,
-      "loss": 0.1039,
+      "epoch": 4.13,
+      "grad_norm": 6.33543586730957,
+      "learning_rate": 1.5870736086175945e-05,
+      "loss": 0.0934,
       "step": 2300
     },
     {
-      "epoch": 4.54,
-      "grad_norm": 2.495684862136841,
-      "learning_rate": 1.5463137996219284e-05,
-      "loss": 0.0886,
+      "epoch": 4.31,
+      "grad_norm": 10.25698184967041,
+      "learning_rate": 1.569120287253142e-05,
+      "loss": 0.068,
       "step": 2400
     },
     {
-      "epoch": 4.73,
-      "grad_norm": 3.7909159660339355,
-      "learning_rate": 1.5274102079395087e-05,
-      "loss": 0.0766,
+      "epoch": 4.49,
+      "grad_norm": 0.05421575903892517,
+      "learning_rate": 1.5511669658886893e-05,
+      "loss": 0.0767,
       "step": 2500
     },
     {
-      "epoch": 4.91,
-      "grad_norm": 0.025901077315211296,
-      "learning_rate": 1.5085066162570889e-05,
-      "loss": 0.0916,
+      "epoch": 4.67,
+      "grad_norm": 0.04917303845286369,
+      "learning_rate": 1.533213644524237e-05,
+      "loss": 0.1053,
       "step": 2600
     },
     {
-      "epoch": 5.0,
-      "eval_loss": 0.3040069341659546,
-      "eval_runtime": 4.2484,
-      "eval_samples_per_second": 91.564,
-      "eval_steps_per_second": 6.591,
-      "step": 2645
-    },
-    {
-      "epoch": 5.1,
-      "grad_norm": 0.011547481641173363,
-      "learning_rate": 1.4896030245746694e-05,
-      "loss": 0.0644,
+      "epoch": 4.85,
+      "grad_norm": 1.722901463508606,
+      "learning_rate": 1.5152603231597847e-05,
+      "loss": 0.0513,
       "step": 2700
     },
     {
-      "epoch": 5.29,
-      "grad_norm": 0.026049258187413216,
-      "learning_rate": 1.4706994328922497e-05,
-      "loss": 0.0542,
+      "epoch": 5.03,
+      "grad_norm": 28.548158645629883,
+      "learning_rate": 1.4973070017953321e-05,
+      "loss": 0.0611,
       "step": 2800
     },
     {
-      "epoch": 5.48,
-      "grad_norm": 0.35288381576538086,
-      "learning_rate": 1.45179584120983e-05,
-      "loss": 0.0512,
+      "epoch": 5.21,
+      "grad_norm": 1.0562294721603394,
+      "learning_rate": 1.4793536804308799e-05,
+      "loss": 0.0543,
       "step": 2900
     },
     {
-      "epoch": 5.67,
-      "grad_norm": 0.023430563509464264,
-      "learning_rate": 1.4328922495274103e-05,
+      "epoch": 5.39,
+      "grad_norm": 0.011326675303280354,
+      "learning_rate": 1.4614003590664274e-05,
       "loss": 0.0491,
       "step": 3000
     },
     {
-      "epoch": 5.86,
-      "grad_norm": 0.0032355256844311953,
-      "learning_rate": 1.4139886578449906e-05,
-      "loss": 0.0326,
+      "epoch": 5.57,
+      "grad_norm": 0.009987740777432919,
+      "learning_rate": 1.4434470377019749e-05,
+      "loss": 0.0567,
       "step": 3100
     },
     {
-      "epoch": 6.0,
-      "eval_loss": 0.3605019450187683,
-      "eval_runtime": 4.2462,
-      "eval_samples_per_second": 91.611,
-      "eval_steps_per_second": 6.594,
-      "step": 3174
-    },
-    {
-      "epoch": 6.05,
-      "grad_norm": 0.1259104311466217,
-      "learning_rate": 1.395085066162571e-05,
-      "loss": 0.0665,
+      "epoch": 5.75,
+      "grad_norm": 26.81354331970215,
+      "learning_rate": 1.4254937163375226e-05,
+      "loss": 0.0641,
       "step": 3200
     },
     {
-      "epoch": 6.24,
-      "grad_norm": 0.08823427557945251,
-      "learning_rate": 1.3761814744801514e-05,
-      "loss": 0.0252,
+      "epoch": 5.92,
+      "grad_norm": 0.042816389352083206,
+      "learning_rate": 1.4075403949730702e-05,
+      "loss": 0.0488,
       "step": 3300
-    },
-    {
-      "epoch": 6.43,
-      "grad_norm": 101.22339630126953,
-      "learning_rate": 1.3572778827977318e-05,
-      "loss": 0.0416,
-      "step": 3400
-    },
-    {
-      "epoch": 6.62,
-      "grad_norm": 0.033835213631391525,
-      "learning_rate": 1.338374291115312e-05,
-      "loss": 0.0465,
-      "step": 3500
-    },
-    {
-      "epoch": 6.81,
-      "grad_norm": 0.15451325476169586,
-      "learning_rate": 1.3194706994328923e-05,
-      "loss": 0.0358,
-      "step": 3600
-    },
-    {
-      "epoch": 6.99,
-      "grad_norm": 0.0031314522493630648,
-      "learning_rate": 1.3005671077504726e-05,
-      "loss": 0.0344,
-      "step": 3700
-    },
-    {
-      "epoch": 7.0,
-      "eval_loss": 0.41286924481391907,
-      "eval_runtime": 4.2519,
-      "eval_samples_per_second": 91.488,
-      "eval_steps_per_second": 6.585,
-      "step": 3703
-    },
-    {
-      "epoch": 7.18,
-      "grad_norm": 0.0014451502356678247,
-      "learning_rate": 1.281663516068053e-05,
-      "loss": 0.0148,
-      "step": 3800
-    },
-    {
-      "epoch": 7.37,
-      "grad_norm": 17.679460525512695,
-      "learning_rate": 1.2627599243856335e-05,
-      "loss": 0.0309,
-      "step": 3900
-    },
-    {
-      "epoch": 7.56,
-      "grad_norm": 0.03577155992388725,
-      "learning_rate": 1.2438563327032138e-05,
-      "loss": 0.0235,
-      "step": 4000
-    },
-    {
-      "epoch": 7.75,
-      "grad_norm": 28.639484405517578,
-      "learning_rate": 1.224952741020794e-05,
-      "loss": 0.0296,
-      "step": 4100
-    },
-    {
-      "epoch": 7.94,
-      "grad_norm": 0.00975144561380148,
-      "learning_rate": 1.2060491493383744e-05,
-      "loss": 0.0218,
-      "step": 4200
-    },
-    {
-      "epoch": 8.0,
-      "eval_loss": 0.3556581735610962,
-      "eval_runtime": 4.245,
-      "eval_samples_per_second": 91.637,
-      "eval_steps_per_second": 6.596,
-      "step": 4232
     }
   ],
   "logging_steps": 100,
-  "max_steps": 10580,
+  "max_steps": 11140,
   "num_input_tokens_seen": 0,
   "num_train_epochs": 20,
   "save_steps": 500,
-  "total_flos": 4.321523491876685e+16,
+  "total_flos": 3.4115676633458784e+16,
   "train_batch_size": 14,
   "trial_name": null,
   "trial_params": null

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e38d84d4a8a34ed1c3d4c1693843cfd19f859389604ec942e9d423aad70c82fc
+oid sha256:491d871fc503f6c09f6650ccce3a497d9ac84504786a77113305d7344d9a5d08
 size 4856