Training in progress, step 3500, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c8e4a377d9d9b4ee3182c91d09c6aa160ff1ade127f1e240b0ace4f4312419ec
+oid sha256:5f249b4c12d0314b83435dfe63816bee13e5910dc2b4ef014ba7a61e89f43ac0
 size 891558696

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:37c76f78e4bea0bb233f5490f2342fc733388c7761a0fa0c0e5fdf8f1a5336d2
+oid sha256:d2341c66cd39e43b1d806239932be769b677cdef8e280c9b4ff34c8958b0ad34
 size 1783272762

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f180d57072cebd56f1e36f710e6b62868e2b14fe85aee7effc0a0d28a6763011
+oid sha256:dbe26b37d45ea8d0357f9b26f439f5ad172b1a279f1b8765178fa166fce80cbc
 size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:19fa64c0f058dbaac84e2a6129da56913abd2f29f4a3f61f13f6abfb2cd3ff5f
+oid sha256:c9c9a3885ae8ac27be8fa78a1d765dfbf434202614784e058d9c43f9ea39114b
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 0.08551913499832153,
-  "best_model_checkpoint": "./fine-tuned/checkpoint-3000",
-  "epoch": 0.48,
+  "best_metric": 0.08540560305118561,
+  "best_model_checkpoint": "./fine-tuned/checkpoint-3500",
+  "epoch": 0.56,
   "eval_steps": 500,
-  "global_step": 3000,
+  "global_step": 3500,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -81,9 +81,9 @@
     {
       "epoch": 0.08,
       "eval_loss": 0.09235642850399017,
-      "eval_runtime": 116.7651,
-      "eval_samples_per_second": 17.128,
-      "eval_steps_per_second": 2.141,
+      "eval_runtime": 109.274,
+      "eval_samples_per_second": 18.303,
+      "eval_steps_per_second": 2.288,
       "step": 500
     },
     {
@@ -159,9 +159,9 @@
     {
       "epoch": 0.16,
       "eval_loss": 0.08949962258338928,
-      "eval_runtime": 116.7407,
-      "eval_samples_per_second": 17.132,
-      "eval_steps_per_second": 2.141,
+      "eval_runtime": 109.2536,
+      "eval_samples_per_second": 18.306,
+      "eval_steps_per_second": 2.288,
       "step": 1000
     },
     {
@@ -237,9 +237,9 @@
     {
       "epoch": 0.24,
       "eval_loss": 0.08808805048465729,
-      "eval_runtime": 116.8722,
-      "eval_samples_per_second": 17.113,
-      "eval_steps_per_second": 2.139,
+      "eval_runtime": 109.2355,
+      "eval_samples_per_second": 18.309,
+      "eval_steps_per_second": 2.289,
       "step": 1500
     },
     {
@@ -315,9 +315,9 @@
     {
       "epoch": 0.32,
       "eval_loss": 0.08704760670661926,
-      "eval_runtime": 116.8362,
-      "eval_samples_per_second": 17.118,
-      "eval_steps_per_second": 2.14,
+      "eval_runtime": 109.4348,
+      "eval_samples_per_second": 18.276,
+      "eval_steps_per_second": 2.284,
       "step": 2000
     },
     {
@@ -393,9 +393,9 @@
     {
       "epoch": 0.4,
       "eval_loss": 0.08615937829017639,
-      "eval_runtime": 116.9591,
-      "eval_samples_per_second": 17.1,
-      "eval_steps_per_second": 2.137,
+      "eval_runtime": 109.2621,
+      "eval_samples_per_second": 18.305,
+      "eval_steps_per_second": 2.288,
       "step": 2500
     },
     {
@@ -471,10 +471,88 @@
     {
       "epoch": 0.48,
       "eval_loss": 0.08551913499832153,
-      "eval_runtime": 116.545,
-      "eval_samples_per_second": 17.161,
-      "eval_steps_per_second": 2.145,
+      "eval_runtime": 109.2626,
+      "eval_samples_per_second": 18.305,
+      "eval_steps_per_second": 2.288,
       "step": 3000
+    },
+    {
+      "epoch": 0.488,
+      "grad_norm": 7060.21826171875,
+      "learning_rate": 2.5120000000000003e-05,
+      "loss": 0.0684,
+      "step": 3050
+    },
+    {
+      "epoch": 0.496,
+      "grad_norm": 7841.55322265625,
+      "learning_rate": 2.504e-05,
+      "loss": 0.0653,
+      "step": 3100
+    },
+    {
+      "epoch": 0.504,
+      "grad_norm": 5290.3271484375,
+      "learning_rate": 2.4959999999999998e-05,
+      "loss": 0.0668,
+      "step": 3150
+    },
+    {
+      "epoch": 0.512,
+      "grad_norm": 6200.4853515625,
+      "learning_rate": 2.4880000000000002e-05,
+      "loss": 0.0665,
+      "step": 3200
+    },
+    {
+      "epoch": 0.52,
+      "grad_norm": 6859.83544921875,
+      "learning_rate": 2.48e-05,
+      "loss": 0.0678,
+      "step": 3250
+    },
+    {
+      "epoch": 0.528,
+      "grad_norm": 7718.70068359375,
+      "learning_rate": 2.472e-05,
+      "loss": 0.0679,
+      "step": 3300
+    },
+    {
+      "epoch": 0.536,
+      "grad_norm": 10752.4873046875,
+      "learning_rate": 2.464e-05,
+      "loss": 0.062,
+      "step": 3350
+    },
+    {
+      "epoch": 0.544,
+      "grad_norm": 6991.5087890625,
+      "learning_rate": 2.456e-05,
+      "loss": 0.0659,
+      "step": 3400
+    },
+    {
+      "epoch": 0.552,
+      "grad_norm": 6204.99658203125,
+      "learning_rate": 2.448e-05,
+      "loss": 0.0636,
+      "step": 3450
+    },
+    {
+      "epoch": 0.56,
+      "grad_norm": 13521.5908203125,
+      "learning_rate": 2.44e-05,
+      "loss": 0.0671,
+      "step": 3500
+    },
+    {
+      "epoch": 0.56,
+      "eval_loss": 0.08540560305118561,
+      "eval_runtime": 109.3641,
+      "eval_samples_per_second": 18.288,
+      "eval_steps_per_second": 2.286,
+      "step": 3500
     }
   ],
   "logging_steps": 50,
@@ -494,7 +572,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 1.461498937344e+16,
+  "total_flos": 1.705082093568e+16,
   "train_batch_size": 8,
   "trial_name": null,
   "trial_params": null