Training in progress, step 6000, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:127785b687ae550620c07c76161ce4fde314a3f786bb3bdecdd030f4cff25442
+oid sha256:fb514424646e7839682dd9dae8b997e057d871af06a5e7e929271aaaa2e7c88d
 size 577789320

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f56a73e3c0f71225db11233195f56bf1f4661b8840ebb310a361c5bbb1b1c79b
+oid sha256:4fb4cccce2cf2fb4228373ebd75bc513ed9c699a9a015d27c99253ddaeab739f
 size 1155772233

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e814e69ac405112ac0e823a174f61a291238359cb5e185b2802f269183d97fb3
+oid sha256:8b78a35397517539ceb5abaec4c078472043c61c90e9313f43ee762be5908798
 size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1f4ebc94f219234dc117648ea36b52869f5ad7c506a78e36de2ec35324b116e9
+oid sha256:cdc16f96a129b2f8ebddb713f251b55486ac014357ce4aedef1d542306c34e74
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 0.3729405403137207,
-  "best_model_checkpoint": "mikhail-panzo/zlm_b128_le5_s8000/checkpoint-5500",
-  "epoch": 9.214659685863875,
+  "best_metric": 0.3696598410606384,
+  "best_model_checkpoint": "mikhail-panzo/zlm_b128_le5_s8000/checkpoint-6000",
+  "epoch": 10.052356020942408,
   "eval_steps": 500,
-  "global_step": 5500,
+  "global_step": 6000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -865,6 +865,84 @@
       "eval_samples_per_second": 32.812,
       "eval_steps_per_second": 4.105,
       "step": 5500
+    },
+    {
+      "epoch": 9.298429319371728,
+      "grad_norm": 1.0570714473724365,
+      "learning_rate": 4.088333333333334e-06,
+      "loss": 0.4144,
+      "step": 5550
+    },
+    {
+      "epoch": 9.38219895287958,
+      "grad_norm": 1.1278653144836426,
+      "learning_rate": 4.005000000000001e-06,
+      "loss": 0.4154,
+      "step": 5600
+    },
+    {
+      "epoch": 9.465968586387435,
+      "grad_norm": 5.811945915222168,
+      "learning_rate": 3.921666666666667e-06,
+      "loss": 0.4085,
+      "step": 5650
+    },
+    {
+      "epoch": 9.549738219895287,
+      "grad_norm": 1.6973522901535034,
+      "learning_rate": 3.8383333333333336e-06,
+      "loss": 0.4133,
+      "step": 5700
+    },
+    {
+      "epoch": 9.633507853403142,
+      "grad_norm": 1.209333062171936,
+      "learning_rate": 3.7550000000000005e-06,
+      "loss": 0.4123,
+      "step": 5750
+    },
+    {
+      "epoch": 9.717277486910994,
+      "grad_norm": 3.592991590499878,
+      "learning_rate": 3.6716666666666665e-06,
+      "loss": 0.4126,
+      "step": 5800
+    },
+    {
+      "epoch": 9.801047120418849,
+      "grad_norm": 1.152239203453064,
+      "learning_rate": 3.588333333333334e-06,
+      "loss": 0.4115,
+      "step": 5850
+    },
+    {
+      "epoch": 9.884816753926701,
+      "grad_norm": 1.6118751764297485,
+      "learning_rate": 3.505e-06,
+      "loss": 0.4036,
+      "step": 5900
+    },
+    {
+      "epoch": 9.968586387434556,
+      "grad_norm": 1.4384329319000244,
+      "learning_rate": 3.4216666666666672e-06,
+      "loss": 0.4103,
+      "step": 5950
+    },
+    {
+      "epoch": 10.052356020942408,
+      "grad_norm": 1.447549819946289,
+      "learning_rate": 3.3383333333333333e-06,
+      "loss": 0.4056,
+      "step": 6000
+    },
+    {
+      "epoch": 10.052356020942408,
+      "eval_loss": 0.3696598410606384,
+      "eval_runtime": 256.5857,
+      "eval_samples_per_second": 33.084,
+      "eval_steps_per_second": 4.139,
+      "step": 6000
     }
   ],
   "logging_steps": 50,
@@ -884,7 +962,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 9.854783789898643e+16,
+  "total_flos": 1.0748974547355264e+17,
   "train_batch_size": 16,
   "trial_name": null,
   "trial_params": null