Training in progress, step 7000, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c1caf9f1f88fe44200f0109ef94036e80d461ce19b24f4f0bd4876dbe777e923
+oid sha256:bc111bfdca8ed66f8f79b1bd7bfe63b080b3cd0c7813c6a04d18b038095b2076
 size 150625560

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4068ffb4d3758e775c1a9defa029bef0ee2704e1a030885062d27923342c7485
+oid sha256:012e342ea6ebe917c7ae83cdc62c745f971ca211463bb0b9f52f2a90259f7532
 size 602335994

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fa90ad2b309f532962514f4faece20cc26bddf7f653b06dc572ffee5bcd113ac
+oid sha256:b7086e9014350d6181c560b7f34a0e20e6a473f5b7d4ab3f99a9989189826cf1
 size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:df3390e9a2d585410c5108433534a385ee7eeb522bbd6ab3b6fa3aad2ff13812
+oid sha256:2eb03be752bb38e04dcd3624dd94e7f08bf42e1daf7a9f0b1188fbafdad08914
 size 1064

last-checkpoint/trainer_state.json

@@ -2,9 +2,9 @@
   "best_global_step": null,
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 1.4483003153431808,
+  "epoch": 1.6897113629162455,
   "eval_steps": 500,
-  "global_step": 6000,
+  "global_step": 7000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -1316,11 +1316,229 @@
       "eval_steps_per_second": 20.473,
       "num_input_tokens_seen": 2898379392,
       "step": 6000
+    },
+    {
+      "epoch": 1.460370867721834,
+      "grad_norm": 0.2734375,
+      "learning_rate": 3.9678587596197375e-05,
+      "loss": 2.1105,
+      "mean_token_accuracy": 0.5535502586700022,
+      "num_input_tokens_seen": 2922428832,
+      "num_tokens": 1231890670.0,
+      "step": 6050
+    },
+    {
+      "epoch": 1.4724414201004874,
+      "grad_norm": 0.2412109375,
+      "learning_rate": 3.948996529349631e-05,
+      "loss": 2.0925,
+      "mean_token_accuracy": 0.5552064320072532,
+      "num_input_tokens_seen": 2946628480,
+      "num_tokens": 1242047017.0,
+      "step": 6100
+    },
+    {
+      "epoch": 1.4845119724791407,
+      "grad_norm": 0.2392578125,
+      "learning_rate": 3.930134299079523e-05,
+      "loss": 2.0991,
+      "mean_token_accuracy": 0.5542361034452915,
+      "num_input_tokens_seen": 2970772896,
+      "num_tokens": 1252270839.0,
+      "step": 6150
+    },
+    {
+      "epoch": 1.4965825248577938,
+      "grad_norm": 0.248046875,
+      "learning_rate": 3.9112720688094164e-05,
+      "loss": 2.0975,
+      "mean_token_accuracy": 0.553666141666472,
+      "num_input_tokens_seen": 2995178928,
+      "num_tokens": 1262575010.0,
+      "step": 6200
+    },
+    {
+      "epoch": 1.508653077236447,
+      "grad_norm": 0.27734375,
+      "learning_rate": 3.8924098385393096e-05,
+      "loss": 2.1016,
+      "mean_token_accuracy": 0.5535378622636199,
+      "num_input_tokens_seen": 3019494064,
+      "num_tokens": 1272716330.0,
+      "step": 6250
+    },
+    {
+      "epoch": 1.5207236296151003,
+      "grad_norm": 0.251953125,
+      "learning_rate": 3.873547608269202e-05,
+      "loss": 2.1071,
+      "mean_token_accuracy": 0.5533343946188688,
+      "num_input_tokens_seen": 3043731184,
+      "num_tokens": 1282959106.0,
+      "step": 6300
+    },
+    {
+      "epoch": 1.5327941819937534,
+      "grad_norm": 0.267578125,
+      "learning_rate": 3.854685377999095e-05,
+      "loss": 2.1057,
+      "mean_token_accuracy": 0.5532900895178318,
+      "num_input_tokens_seen": 3067867536,
+      "num_tokens": 1293166946.0,
+      "step": 6350
+    },
+    {
+      "epoch": 1.5448647343724067,
+      "grad_norm": 0.32421875,
+      "learning_rate": 3.835823147728987e-05,
+      "loss": 2.106,
+      "mean_token_accuracy": 0.5535468808189035,
+      "num_input_tokens_seen": 3092176096,
+      "num_tokens": 1303381117.0,
+      "step": 6400
+    },
+    {
+      "epoch": 1.55693528675106,
+      "grad_norm": 0.248046875,
+      "learning_rate": 3.8169609174588804e-05,
+      "loss": 2.0989,
+      "mean_token_accuracy": 0.5537711648643017,
+      "num_input_tokens_seen": 3116219440,
+      "num_tokens": 1313584312.0,
+      "step": 6450
+    },
+    {
+      "epoch": 1.569005839129713,
+      "grad_norm": 0.2490234375,
+      "learning_rate": 3.7980986871887736e-05,
+      "loss": 2.1121,
+      "num_input_tokens_seen": 3140306656,
+      "step": 6500
+    },
+    {
+      "epoch": 1.569005839129713,
+      "eval_loss": 1.972907304763794,
+      "eval_mean_token_accuracy": 0.5776848248619922,
+      "eval_num_tokens": 1323681372.0,
+      "eval_runtime": 129.9521,
+      "eval_samples_per_second": 82.43,
+      "eval_steps_per_second": 20.608,
+      "num_input_tokens_seen": 3140306656,
+      "step": 6500
+    },
+    {
+      "epoch": 1.5810763915083665,
+      "grad_norm": 0.255859375,
+      "learning_rate": 3.779236456918666e-05,
+      "loss": 2.1022,
+      "mean_token_accuracy": 0.5528610655851662,
+      "num_input_tokens_seen": 3164451568,
+      "num_tokens": 1333870621.0,
+      "step": 6550
+    },
+    {
+      "epoch": 1.5931469438870196,
+      "grad_norm": 0.2578125,
+      "learning_rate": 3.760374226648559e-05,
+      "loss": 2.0995,
+      "mean_token_accuracy": 0.5544479803740978,
+      "num_input_tokens_seen": 3188556560,
+      "num_tokens": 1344040900.0,
+      "step": 6600
+    },
+    {
+      "epoch": 1.605217496265673,
+      "grad_norm": 0.263671875,
+      "learning_rate": 3.741511996378452e-05,
+      "loss": 2.1007,
+      "mean_token_accuracy": 0.5533806948363781,
+      "num_input_tokens_seen": 3212749536,
+      "num_tokens": 1354214136.0,
+      "step": 6650
+    },
+    {
+      "epoch": 1.6172880486443262,
+      "grad_norm": 0.314453125,
+      "learning_rate": 3.722649766108345e-05,
+      "loss": 2.1034,
+      "mean_token_accuracy": 0.5535299601778388,
+      "num_input_tokens_seen": 3236971856,
+      "num_tokens": 1364392553.0,
+      "step": 6700
+    },
+    {
+      "epoch": 1.6293586010229792,
+      "grad_norm": 0.255859375,
+      "learning_rate": 3.7037875358382376e-05,
+      "loss": 2.1039,
+      "mean_token_accuracy": 0.5538195591047406,
+      "num_input_tokens_seen": 3261221664,
+      "num_tokens": 1374597748.0,
+      "step": 6750
+    },
+    {
+      "epoch": 1.6414291534016325,
+      "grad_norm": 0.2431640625,
+      "learning_rate": 3.68492530556813e-05,
+      "loss": 2.0999,
+      "mean_token_accuracy": 0.5533070769160986,
+      "num_input_tokens_seen": 3285325344,
+      "num_tokens": 1384799126.0,
+      "step": 6800
+    },
+    {
+      "epoch": 1.6534997057802858,
+      "grad_norm": 0.294921875,
+      "learning_rate": 3.666063075298023e-05,
+      "loss": 2.0969,
+      "mean_token_accuracy": 0.5540298366174102,
+      "num_input_tokens_seen": 3309447808,
+      "num_tokens": 1394991541.0,
+      "step": 6850
+    },
+    {
+      "epoch": 1.6655702581589389,
+      "grad_norm": 0.28125,
+      "learning_rate": 3.647200845027916e-05,
+      "loss": 2.0982,
+      "mean_token_accuracy": 0.5548456938192249,
+      "num_input_tokens_seen": 3333810384,
+      "num_tokens": 1405193552.0,
+      "step": 6900
+    },
+    {
+      "epoch": 1.6776408105375922,
+      "grad_norm": 0.27734375,
+      "learning_rate": 3.628338614757809e-05,
+      "loss": 2.0998,
+      "mean_token_accuracy": 0.5537503241375089,
+      "num_input_tokens_seen": 3357917184,
+      "num_tokens": 1415265616.0,
+      "step": 6950
+    },
+    {
+      "epoch": 1.6897113629162455,
+      "grad_norm": 0.236328125,
+      "learning_rate": 3.609476384487702e-05,
+      "loss": 2.0829,
+      "num_input_tokens_seen": 3382174368,
+      "step": 7000
+    },
+    {
+      "epoch": 1.6897113629162455,
+      "eval_loss": 1.971500039100647,
+      "eval_mean_token_accuracy": 0.5780662869320956,
+      "eval_num_tokens": 1425356084.0,
+      "eval_runtime": 130.448,
+      "eval_samples_per_second": 82.117,
+      "eval_steps_per_second": 20.529,
+      "num_input_tokens_seen": 3382174368,
+      "step": 7000
     }
   ],
   "logging_steps": 50,
   "max_steps": 16568,
-  "num_input_tokens_seen": 2898379392,
+  "num_input_tokens_seen": 3382174368,
   "num_train_epochs": 4,
   "save_steps": 1000,
   "stateful_callbacks": {
@@ -1335,7 +1553,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 7.753447755428659e+17,
+  "total_flos": 9.047646534618317e+17,
   "train_batch_size": 16,
   "trial_name": null,
   "trial_params": null