Training in progress, step 3900, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bae5f432ff85d88eb29fafe392aae70b0acdb17f7a92f02a4494c4c316f32e83
+oid sha256:4976734d9102aa4ee192bfa84e0b98f4d622b4c2ecd3bb1d2bb533f92a4a0e7e
 size 75507072

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:74f7e65012cc8537c1d12565e9cda03c555c0dd8d47b19f728f605d9f2f75fdb
+oid sha256:f007b8cd0bc78612497a01e58cda162ea087c6bd751572cc89b5116a21cff436
 size 151032837

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0b234d9d700f882ce9920f729494e9162825344c4091455e69b9d19ef11133c1
+oid sha256:f1b836b0763d8c8c7392ab9bf480880e0c4fd0fbff6b82477d0ff13a6bddc661
 size 14575

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3961d2f201ce99426ee113655da78d8214f6ef72fc082855daa3c565989888c2
+oid sha256:a42424f90eb68391066979b79f87ab18e2fce2a0abc640b46cf1b3f4c12c720a
 size 627

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 1.116302251815796,
-  "best_model_checkpoint": "./outputs/checkpoint-3800",
-  "epoch": 2.769398907103825,
+  "best_metric": 1.1058732271194458,
+  "best_model_checkpoint": "./outputs/checkpoint-3900",
+  "epoch": 2.8422586520947175,
   "eval_steps": 100,
-  "global_step": 3800,
+  "global_step": 3900,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -539,13 +539,27 @@
       "eval_samples_per_second": 18.455,
       "eval_steps_per_second": 2.309,
       "step": 3800
+    },
+    {
+      "epoch": 2.84,
+      "learning_rate": 0.0002,
+      "loss": 1.0902,
+      "step": 3900
+    },
+    {
+      "epoch": 2.84,
+      "eval_loss": 1.1058732271194458,
+      "eval_runtime": 339.729,
+      "eval_samples_per_second": 18.468,
+      "eval_steps_per_second": 2.311,
+      "step": 3900
     }
   ],
   "logging_steps": 100,
   "max_steps": 4116,
   "num_train_epochs": 3,
   "save_steps": 100,
-  "total_flos": 1.2320741436517632e+18,
+  "total_flos": 1.2644892278408356e+18,
   "trial_name": null,
   "trial_params": null
 }