Training in progress, step 20000, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d69e130591023d3408c94e3445de9f9023a881b14d399648ae4226ee5bff553b
+oid sha256:80f6cba2576f1ee2628aa99fb4980b692c7f23dc4a1cc36b06da1ee070ed4750
 size 496262784

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a86f5fd274e719671c5f2463396daaf1a7a1c0ce3b5b26f63b7d962eb3bc34f1
+oid sha256:238c57d596e4b481c83fa003f41489533d0e0d6fa65bc29cfff61620056f3f43
 size 992621963

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7a721edff9706126df51d3abc78a93078e6b8e75513d77f55a62d72e4c5b16cf
+oid sha256:f2d3bdd032f891e90f3fae19dbb6eb0bfaf6ef591769711ba856b31e40586a5e
 size 14645

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5637b5a3d72a0623ba9410834e3d0160dfe8e69a524a3dc9973a65396f79b127
+oid sha256:c5b7f9f4f6c99c7a7ebbb7780b490f0a9830b3a79c1c5a07331e7e6edccccd28
 size 1465

last-checkpoint/trainer_state.json

@@ -2,9 +2,9 @@
   "best_global_step": null,
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.36,
+  "epoch": 0.4,
   "eval_steps": 1000,
-  "global_step": 18000,
+  "global_step": 20000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -1430,6 +1430,164 @@
       "eval_samples_per_second": 306.051,
       "eval_steps_per_second": 19.237,
       "step": 18000
+    },
+    {
+      "epoch": 0.362,
+      "grad_norm": 0.2838000953197479,
+      "learning_rate": 0.0004276824644603691,
+      "loss": 3.5090277099609377,
+      "step": 18100
+    },
+    {
+      "epoch": 0.364,
+      "grad_norm": 0.6395978331565857,
+      "learning_rate": 0.0004259690972040426,
+      "loss": 3.5145654296875,
+      "step": 18200
+    },
+    {
+      "epoch": 0.366,
+      "grad_norm": 0.1982821524143219,
+      "learning_rate": 0.0004242507269304744,
+      "loss": 3.504261474609375,
+      "step": 18300
+    },
+    {
+      "epoch": 0.368,
+      "grad_norm": 0.22082549333572388,
+      "learning_rate": 0.0004225274218868482,
+      "loss": 3.496566162109375,
+      "step": 18400
+    },
+    {
+      "epoch": 0.37,
+      "grad_norm": 0.2385374903678894,
+      "learning_rate": 0.0004207992505163379,
+      "loss": 3.514676208496094,
+      "step": 18500
+    },
+    {
+      "epoch": 0.372,
+      "grad_norm": 0.18938113749027252,
+      "learning_rate": 0.00041906628145538987,
+      "loss": 3.5016729736328127,
+      "step": 18600
+    },
+    {
+      "epoch": 0.374,
+      "grad_norm": 0.17800572514533997,
+      "learning_rate": 0.0004173285835309965,
+      "loss": 3.471422119140625,
+      "step": 18700
+    },
+    {
+      "epoch": 0.376,
+      "grad_norm": 0.19377586245536804,
+      "learning_rate": 0.0004155862257579626,
+      "loss": 3.5025115966796876,
+      "step": 18800
+    },
+    {
+      "epoch": 0.378,
+      "grad_norm": 0.21612316370010376,
+      "learning_rate": 0.00041383927733616477,
+      "loss": 3.4965274047851564,
+      "step": 18900
+    },
+    {
+      "epoch": 0.38,
+      "grad_norm": 0.6052186489105225,
+      "learning_rate": 0.000412087807647803,
+      "loss": 3.4777236938476563,
+      "step": 19000
+    },
+    {
+      "epoch": 0.38,
+      "eval_accuracy": 0.3683073364850164,
+      "eval_loss": 3.4274120330810547,
+      "eval_runtime": 6.5604,
+      "eval_samples_per_second": 295.864,
+      "eval_steps_per_second": 18.596,
+      "step": 19000
+    },
+    {
+      "epoch": 0.382,
+      "grad_norm": 0.19751021265983582,
+      "learning_rate": 0.0004103318862546447,
+      "loss": 3.4927557373046874,
+      "step": 19100
+    },
+    {
+      "epoch": 0.384,
+      "grad_norm": 0.295538067817688,
+      "learning_rate": 0.0004085715828952621,
+      "loss": 3.4930191040039062,
+      "step": 19200
+    },
+    {
+      "epoch": 0.386,
+      "grad_norm": 0.18652473390102386,
+      "learning_rate": 0.00040680696748226293,
+      "loss": 3.4893539428710936,
+      "step": 19300
+    },
+    {
+      "epoch": 0.388,
+      "grad_norm": 0.19307976961135864,
+      "learning_rate": 0.00040503811009951325,
+      "loss": 3.484967041015625,
+      "step": 19400
+    },
+    {
+      "epoch": 0.39,
+      "grad_norm": 0.20981250703334808,
+      "learning_rate": 0.0004032650809993542,
+      "loss": 3.4780728149414064,
+      "step": 19500
+    },
+    {
+      "epoch": 0.392,
+      "grad_norm": 0.31385481357574463,
+      "learning_rate": 0.00040148795059981155,
+      "loss": 3.496181640625,
+      "step": 19600
+    },
+    {
+      "epoch": 0.394,
+      "grad_norm": 0.1942823827266693,
+      "learning_rate": 0.0003997067894817995,
+      "loss": 3.487859802246094,
+      "step": 19700
+    },
+    {
+      "epoch": 0.396,
+      "grad_norm": 0.1936560422182083,
+      "learning_rate": 0.0003979216683863172,
+      "loss": 3.479795227050781,
+      "step": 19800
+    },
+    {
+      "epoch": 0.398,
+      "grad_norm": 0.2430066615343094,
+      "learning_rate": 0.00039613265821163883,
+      "loss": 3.4654913330078125,
+      "step": 19900
+    },
+    {
+      "epoch": 0.4,
+      "grad_norm": 0.20538495481014252,
+      "learning_rate": 0.0003943398300104985,
+      "loss": 3.4708984375,
+      "step": 20000
+    },
+    {
+      "epoch": 0.4,
+      "eval_accuracy": 0.36907962990408844,
+      "eval_loss": 3.415916919708252,
+      "eval_runtime": 5.8841,
+      "eval_samples_per_second": 329.872,
+      "eval_steps_per_second": 20.734,
+      "step": 20000
     }
   ],
   "logging_steps": 100,
@@ -1449,7 +1607,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 5.6439078912e+17,
+  "total_flos": 6.271008768e+17,
   "train_batch_size": 120,
   "trial_name": null,
   "trial_params": null