Training in progress, step 18000, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a7c3e7cfbc9b6b5cbf70eb556e69c0e89a7feac7c489589a3bd75e6d5bb1f38b
+oid sha256:91f77bd1ac2f3fafb2d2dff8abe3cf8b6a15fc5e2c266220cfd41101a08926f2
 size 579824888

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cb3239c55a1abdd9eba611bcd166c3c48906471881b6a03190d17e7f82682db4
+oid sha256:9bd91aa34b703c596d190d43204798100682f3f8c6905d18b1c963356249d128
 size 1159794763

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a1e6ca9f6a71acdc5481b681d658fb168e5db2fd263350651883dffaef0871cf
+oid sha256:2118ba9622183c33aa1d1262712b3091d1f33a50e27c8636e086286e6c061b97
 size 14645

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4cfc12145708b72afe7d410dd5ac1b16484b3ab33d6d903b227818b64d609309
+oid sha256:803e8ad8cb04d8d95ab5fe814114135859d42a7ef5d29f0ea36f7faef8803122
 size 1465

last-checkpoint/trainer_state.json

@@ -2,9 +2,9 @@
   "best_global_step": null,
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.32,
+  "epoch": 0.36,
   "eval_steps": 1000,
-  "global_step": 16000,
+  "global_step": 18000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -1272,6 +1272,164 @@
       "eval_samples_per_second": 308.75,
       "eval_steps_per_second": 19.406,
       "step": 16000
+    },
+    {
+      "epoch": 0.322,
+      "grad_norm": 0.16945767402648926,
+      "learning_rate": 0.0007679932081794532,
+      "loss": 3.5018475341796873,
+      "step": 16100
+    },
+    {
+      "epoch": 0.324,
+      "grad_norm": 0.14158344268798828,
+      "learning_rate": 0.0007653277072568166,
+      "loss": 3.52821044921875,
+      "step": 16200
+    },
+    {
+      "epoch": 0.326,
+      "grad_norm": 0.15132898092269897,
+      "learning_rate": 0.0007626516685187669,
+      "loss": 3.5037826538085937,
+      "step": 16300
+    },
+    {
+      "epoch": 0.328,
+      "grad_norm": 0.1444397270679474,
+      "learning_rate": 0.0007599651982474656,
+      "loss": 3.5029248046875,
+      "step": 16400
+    },
+    {
+      "epoch": 0.33,
+      "grad_norm": 0.15066049993038177,
+      "learning_rate": 0.0007572684031393755,
+      "loss": 3.5109619140625,
+      "step": 16500
+    },
+    {
+      "epoch": 0.332,
+      "grad_norm": 0.15420712530612946,
+      "learning_rate": 0.0007545613903010231,
+      "loss": 3.498372802734375,
+      "step": 16600
+    },
+    {
+      "epoch": 0.334,
+      "grad_norm": 0.15669448673725128,
+      "learning_rate": 0.000751844267244744,
+      "loss": 3.504073181152344,
+      "step": 16700
+    },
+    {
+      "epoch": 0.336,
+      "grad_norm": 0.14457182586193085,
+      "learning_rate": 0.000749117141884414,
+      "loss": 3.5009725952148436,
+      "step": 16800
+    },
+    {
+      "epoch": 0.338,
+      "grad_norm": 0.1625896543264389,
+      "learning_rate": 0.0007463801225311631,
+      "loss": 3.5116473388671876,
+      "step": 16900
+    },
+    {
+      "epoch": 0.34,
+      "grad_norm": 0.1407516449689865,
+      "learning_rate": 0.0007436333178890721,
+      "loss": 3.4775790405273437,
+      "step": 17000
+    },
+    {
+      "epoch": 0.34,
+      "eval_accuracy": 0.3666105090381519,
+      "eval_loss": 3.4458343982696533,
+      "eval_runtime": 6.6119,
+      "eval_samples_per_second": 293.56,
+      "eval_steps_per_second": 18.451,
+      "step": 17000
+    },
+    {
+      "epoch": 0.342,
+      "grad_norm": 0.15488570928573608,
+      "learning_rate": 0.0007408768370508576,
+      "loss": 3.5057693481445313,
+      "step": 17100
+    },
+    {
+      "epoch": 0.344,
+      "grad_norm": 0.13969258964061737,
+      "learning_rate": 0.0007381107894935382,
+      "loss": 3.4655438232421876,
+      "step": 17200
+    },
+    {
+      "epoch": 0.346,
+      "grad_norm": 0.15081284940242767,
+      "learning_rate": 0.0007353352850740858,
+      "loss": 3.490411376953125,
+      "step": 17300
+    },
+    {
+      "epoch": 0.348,
+      "grad_norm": 0.1477256566286087,
+      "learning_rate": 0.0007325504340250635,
+      "loss": 3.48463134765625,
+      "step": 17400
+    },
+    {
+      "epoch": 0.35,
+      "grad_norm": 0.15759490430355072,
+      "learning_rate": 0.0007297563469502479,
+      "loss": 3.5031024169921876,
+      "step": 17500
+    },
+    {
+      "epoch": 0.352,
+      "grad_norm": 0.14924432337284088,
+      "learning_rate": 0.0007269531348202344,
+      "loss": 3.5010330200195314,
+      "step": 17600
+    },
+    {
+      "epoch": 0.354,
+      "grad_norm": 0.21179653704166412,
+      "learning_rate": 0.0007241409089680324,
+      "loss": 3.484757385253906,
+      "step": 17700
+    },
+    {
+      "epoch": 0.356,
+      "grad_norm": 0.1406019777059555,
+      "learning_rate": 0.0007213197810846416,
+      "loss": 3.4941912841796876,
+      "step": 17800
+    },
+    {
+      "epoch": 0.358,
+      "grad_norm": 0.18723711371421814,
+      "learning_rate": 0.0007184898632146171,
+      "loss": 3.4815142822265623,
+      "step": 17900
+    },
+    {
+      "epoch": 0.36,
+      "grad_norm": 0.15316542983055115,
+      "learning_rate": 0.0007156512677516189,
+      "loss": 3.4828509521484374,
+      "step": 18000
+    },
+    {
+      "epoch": 0.36,
+      "eval_accuracy": 0.36743422147076527,
+      "eval_loss": 3.434631824493408,
+      "eval_runtime": 6.7452,
+      "eval_samples_per_second": 287.759,
+      "eval_steps_per_second": 18.087,
+      "step": 18000
     }
   ],
   "logging_steps": 100,
@@ -1291,7 +1449,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 6.2719373869056e+17,
+  "total_flos": 7.0559295602688e+17,
   "train_batch_size": 120,
   "trial_name": null,
   "trial_params": null