Training in progress, step 3200, checkpoint

last-checkpoint/adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:71fd0fa3c83db34c04a261e6bd3242a5476745a9f7511efddbff7145d0dc8239
+oid sha256:7773ef07966a2f8859ef5fb5da82bbb13eb8d04fd139706b6c690c50c47de356
 size 50349441

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e5660c8b38f29f9c62edfdc1f87d539132c89fbba7620bbffe5cb7cb482329b9
+oid sha256:02ac7514df3527aad95b44dfa9c74475ae9fbadaa994bd811ae4a4f4eaa7b9fc
 size 100691721

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6e42983a0941a5ea3ab8dd356cda0bea564cad32d8e894b126d9bb07e48e1e57
+oid sha256:084132876d83624f29d3863ed0c0b0e930fe3b7a99f53b6167286f214214e720
 size 14575

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0f4766a50a63740d51a19c983a4de442d7f62ff1e951fc016c44cc6c58fb9db1
+oid sha256:50b57a34df83b700e2c13775ff734b4569b74ce7e20da3479db76577bb4e906e
 size 627

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 1.320979356765747,
-  "best_model_checkpoint": "./outputs/checkpoint-3100",
-  "epoch": 2.259016393442623,
+  "best_metric": 1.3095366954803467,
+  "best_model_checkpoint": "./outputs/checkpoint-3200",
+  "epoch": 2.3318761384335156,
   "eval_steps": 100,
-  "global_step": 3100,
+  "global_step": 3200,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -441,13 +441,27 @@
       "eval_samples_per_second": 47.034,
       "eval_steps_per_second": 5.885,
       "step": 3100
+    },
+    {
+      "epoch": 2.33,
+      "learning_rate": 0.0002,
+      "loss": 1.2953,
+      "step": 3200
+    },
+    {
+      "epoch": 2.33,
+      "eval_loss": 1.3095366954803467,
+      "eval_runtime": 138.3583,
+      "eval_samples_per_second": 45.346,
+      "eval_steps_per_second": 5.674,
+      "step": 3200
     }
   ],
   "logging_steps": 100,
   "max_steps": 4116,
   "num_train_epochs": 3,
   "save_steps": 100,
-  "total_flos": 1.8479244039539098e+17,
+  "total_flos": 1.9074451823682355e+17,
   "trial_name": null,
   "trial_params": null
 }