Training in progress, step 1200, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:47ede7c685c4cf2a44431641bbe68a34d9f802e8687f0815dd32ac6b485ba701
+oid sha256:04344c1d1217c1158115c3b0929c1cf08a3369d5b27d880cb55a5627a0758da5
 size 14764744

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:84faa572c4eb5bc6dd91c7c16c399ec0f47a06a254293a79c4c40bc096e724b7
+oid sha256:8a50a473d9696eda62ce8fc8006e74e0eacfd768f1e3ac189552cbbf09b3ebc6
 size 29614667

last-checkpoint/rng_state_0.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:787e2ea174d158f4e96070eeff49589d8fa83d2d3249c84689dcdfb8dc3f30e6
+oid sha256:b0dd483bc226b1bbb74ce1edb1ca54fa292eaa7d36ca7dd3fa4f5dc0358a386c
 size 14917

last-checkpoint/rng_state_1.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:64b69f01f980cdba0e54bde1be1ffecd2dd3368d7455083782e53272ee6c7659
+oid sha256:5ab1e3f6b235da786bb29a07f4bd9aff1a3dcf320087ad886f22075df665f7bf
 size 14917

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:14ae2a2128444abab378aa06c09a61a84665f758fcc19fc46f5789b0bc1b5665
+oid sha256:18783150ac09b6b81cea5af47876a10bfe5f36c3d76aca4ffce5382bdfaf7b28
 size 1383

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:458efd6e56e53523ef1da1191c603dc16e272c67a29ec753d749f5877c5d56ac
+oid sha256:c5e22ced5d47ecc95b81d09635f35a609202e47f5e060c6d8daac35d19ea2334
 size 1465

last-checkpoint/trainer_state.json

@@ -2,9 +2,9 @@
   "best_global_step": null,
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.17224303492227533,
+  "epoch": 0.2066916419067304,
   "eval_steps": 100,
-  "global_step": 1000,
+  "global_step": 1200,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -1118,6 +1118,228 @@
       "eval_samples_per_second": 4.527,
       "eval_steps_per_second": 1.132,
       "step": 1000
+    },
+    {
+      "entropy": 0.8927131168544292,
+      "epoch": 0.1739654652714981,
+      "grad_norm": 0.30148547887802124,
+      "learning_rate": 1.6524285222183948e-05,
+      "loss": 0.8974312782287598,
+      "mean_token_accuracy": 0.8002020284533501,
+      "num_tokens": 13944590.0,
+      "step": 1010
+    },
+    {
+      "entropy": 0.8714240893721581,
+      "epoch": 0.17568789562072085,
+      "grad_norm": 0.32693004608154297,
+      "learning_rate": 1.6489838098518775e-05,
+      "loss": 0.8826124191284179,
+      "mean_token_accuracy": 0.8055419594049453,
+      "num_tokens": 14088537.0,
+      "step": 1020
+    },
+    {
+      "entropy": 0.8926309160888195,
+      "epoch": 0.17741032596994358,
+      "grad_norm": 0.32023537158966064,
+      "learning_rate": 1.64553909748536e-05,
+      "loss": 0.9113988876342773,
+      "mean_token_accuracy": 0.7987354248762131,
+      "num_tokens": 14224374.0,
+      "step": 1030
+    },
+    {
+      "entropy": 0.8674004882574081,
+      "epoch": 0.17913275631916634,
+      "grad_norm": 0.28465598821640015,
+      "learning_rate": 1.6420943851188428e-05,
+      "loss": 0.8814109802246094,
+      "mean_token_accuracy": 0.8047246292233468,
+      "num_tokens": 14360301.0,
+      "step": 1040
+    },
+    {
+      "entropy": 0.9156268492341042,
+      "epoch": 0.1808551866683891,
+      "grad_norm": 0.2746152877807617,
+      "learning_rate": 1.6386496727523255e-05,
+      "loss": 0.9329729080200195,
+      "mean_token_accuracy": 0.7970086365938187,
+      "num_tokens": 14497957.0,
+      "step": 1050
+    },
+    {
+      "entropy": 0.8696728102862835,
+      "epoch": 0.18257761701761185,
+      "grad_norm": 0.3130994737148285,
+      "learning_rate": 1.635204960385808e-05,
+      "loss": 0.8681984901428222,
+      "mean_token_accuracy": 0.8039918035268784,
+      "num_tokens": 14636779.0,
+      "step": 1060
+    },
+    {
+      "entropy": 0.9261205658316612,
+      "epoch": 0.1843000473668346,
+      "grad_norm": 0.3177715539932251,
+      "learning_rate": 1.6317602480192905e-05,
+      "loss": 0.9432561874389649,
+      "mean_token_accuracy": 0.79266317486763,
+      "num_tokens": 14775357.0,
+      "step": 1070
+    },
+    {
+      "entropy": 0.8939791619777679,
+      "epoch": 0.18602247771605734,
+      "grad_norm": 0.3282929062843323,
+      "learning_rate": 1.628315535652773e-05,
+      "loss": 0.9121170043945312,
+      "mean_token_accuracy": 0.7999403312802315,
+      "num_tokens": 14914329.0,
+      "step": 1080
+    },
+    {
+      "entropy": 0.9411349311470986,
+      "epoch": 0.1877449080652801,
+      "grad_norm": 0.33822765946388245,
+      "learning_rate": 1.624870823286256e-05,
+      "loss": 0.9574422836303711,
+      "mean_token_accuracy": 0.7887647718191146,
+      "num_tokens": 15050944.0,
+      "step": 1090
+    },
+    {
+      "entropy": 0.9062761172652245,
+      "epoch": 0.18946733841450286,
+      "grad_norm": 0.3344680070877075,
+      "learning_rate": 1.6214261109197382e-05,
+      "loss": 0.9132010459899902,
+      "mean_token_accuracy": 0.7973324447870255,
+      "num_tokens": 15191456.0,
+      "step": 1100
+    },
+    {
+      "epoch": 0.18946733841450286,
+      "eval_entropy": 0.9062020365742001,
+      "eval_loss": 0.9168528914451599,
+      "eval_mean_token_accuracy": 0.7978820896388855,
+      "eval_num_tokens": 15191456.0,
+      "eval_runtime": 2284.6786,
+      "eval_samples_per_second": 4.517,
+      "eval_steps_per_second": 1.13,
+      "step": 1100
+    },
+    {
+      "entropy": 0.8828578740358353,
+      "epoch": 0.19118976876372562,
+      "grad_norm": 0.31287097930908203,
+      "learning_rate": 1.617981398553221e-05,
+      "loss": 0.8892689704895019,
+      "mean_token_accuracy": 0.8019500032067299,
+      "num_tokens": 15329187.0,
+      "step": 1110
+    },
+    {
+      "entropy": 0.8584744602441787,
+      "epoch": 0.19291219911294838,
+      "grad_norm": 0.33894848823547363,
+      "learning_rate": 1.6145366861867035e-05,
+      "loss": 0.8761276245117188,
+      "mean_token_accuracy": 0.8069836989045143,
+      "num_tokens": 15463428.0,
+      "step": 1120
+    },
+    {
+      "entropy": 0.8959764569997788,
+      "epoch": 0.1946346294621711,
+      "grad_norm": 0.335448294878006,
+      "learning_rate": 1.6110919738201862e-05,
+      "loss": 0.914846134185791,
+      "mean_token_accuracy": 0.7979757830500602,
+      "num_tokens": 15598833.0,
+      "step": 1130
+    },
+    {
+      "entropy": 0.8600837841629982,
+      "epoch": 0.19635705981139387,
+      "grad_norm": 0.36168724298477173,
+      "learning_rate": 1.6076472614536685e-05,
+      "loss": 0.874116325378418,
+      "mean_token_accuracy": 0.8072359189391136,
+      "num_tokens": 15735190.0,
+      "step": 1140
+    },
+    {
+      "entropy": 0.8260842703282834,
+      "epoch": 0.19807949016061663,
+      "grad_norm": 0.3406086266040802,
+      "learning_rate": 1.6042025490871512e-05,
+      "loss": 0.846173095703125,
+      "mean_token_accuracy": 0.8120132207870483,
+      "num_tokens": 15873365.0,
+      "step": 1150
+    },
+    {
+      "entropy": 0.8754205822944641,
+      "epoch": 0.1998019205098394,
+      "grad_norm": 0.3090217709541321,
+      "learning_rate": 1.600757836720634e-05,
+      "loss": 0.8730321884155273,
+      "mean_token_accuracy": 0.8036265060305595,
+      "num_tokens": 16012267.0,
+      "step": 1160
+    },
+    {
+      "entropy": 0.8644778996706008,
+      "epoch": 0.20152435085906215,
+      "grad_norm": 0.3449987769126892,
+      "learning_rate": 1.5973131243541165e-05,
+      "loss": 0.8761721611022949,
+      "mean_token_accuracy": 0.8046121463179589,
+      "num_tokens": 16153551.0,
+      "step": 1170
+    },
+    {
+      "entropy": 0.8786791607737541,
+      "epoch": 0.20324678120828488,
+      "grad_norm": 0.32610586285591125,
+      "learning_rate": 1.5938684119875992e-05,
+      "loss": 0.8917497634887696,
+      "mean_token_accuracy": 0.8018891140818596,
+      "num_tokens": 16290644.0,
+      "step": 1180
+    },
+    {
+      "entropy": 0.8934277132153511,
+      "epoch": 0.20496921155750764,
+      "grad_norm": 0.3260601758956909,
+      "learning_rate": 1.590423699621082e-05,
+      "loss": 0.9085002899169922,
+      "mean_token_accuracy": 0.7983957454562187,
+      "num_tokens": 16428870.0,
+      "step": 1190
+    },
+    {
+      "entropy": 0.8712983086705208,
+      "epoch": 0.2066916419067304,
+      "grad_norm": 0.3234138786792755,
+      "learning_rate": 1.5869789872545646e-05,
+      "loss": 0.8846115112304688,
+      "mean_token_accuracy": 0.8027477219700814,
+      "num_tokens": 16570837.0,
+      "step": 1200
+    },
+    {
+      "epoch": 0.2066916419067304,
+      "eval_entropy": 0.9033433407002389,
+      "eval_loss": 0.9155807495117188,
+      "eval_mean_token_accuracy": 0.7982669015902505,
+      "eval_num_tokens": 16570837.0,
+      "eval_runtime": 2278.3123,
+      "eval_samples_per_second": 4.53,
+      "eval_steps_per_second": 1.133,
+      "step": 1200
     }
   ],
   "logging_steps": 10,
@@ -1137,7 +1359,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 1.645082202729349e+17,
+  "total_flos": 1.9741542298157056e+17,
   "train_batch_size": 2,
   "trial_name": null,
   "trial_params": null