Training in progress, step 4000, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:49f9868e4d1cd21d90596284b88521aacdf829dc439af2b6b38673ec9ca76abf
+oid sha256:67fb2641a47847de545739fa9bf331cc86382b9efd2c9bde87633e135625555f
 size 50338848

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8d638a7e1318513481edad79747bbd9d85854414852c1a384c0c1190d66d2abd
+oid sha256:dc3cc469c5109a2b089c90098e5ec399489d763128d9b4b4f6ebef13c7d76fa3
 size 100693001

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:44f7a23b7a631b1fe28c32f80339cec68c940273b7a37ff68a8e64f1b9f09a97
+oid sha256:8993bfd358e6597f1a1d1729c472a0ff6be3dbbb3d5625584e41e187526fa017
 size 14575

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:794a0a668693664bb45fb10418080981a1873b71841f3a55ed72c5aa6d455597
+oid sha256:0ab644d38b64875fe4b9db6a2dd788efbee5d9e5dadfabf0dd125e8dae428442
 size 627

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 1.3290683031082153,
-  "best_model_checkpoint": "./outputs/checkpoint-3900",
-  "epoch": 2.841530054644809,
+  "best_metric": 1.3191989660263062,
+  "best_model_checkpoint": "./outputs/checkpoint-4000",
+  "epoch": 2.9143897996357016,
   "eval_steps": 100,
-  "global_step": 3900,
+  "global_step": 4000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -553,13 +553,27 @@
       "eval_samples_per_second": 43.714,
       "eval_steps_per_second": 5.47,
       "step": 3900
+    },
+    {
+      "epoch": 2.91,
+      "learning_rate": 0.0002,
+      "loss": 1.2978,
+      "step": 4000
+    },
+    {
+      "epoch": 2.91,
+      "eval_loss": 1.3191989660263062,
+      "eval_runtime": 143.6032,
+      "eval_samples_per_second": 43.69,
+      "eval_steps_per_second": 5.466,
+      "step": 4000
     }
   ],
   "logging_steps": 100,
   "max_steps": 4116,
   "num_train_epochs": 3,
   "save_steps": 100,
-  "total_flos": 2.322136123818148e+17,
+  "total_flos": 2.3815364444912026e+17,
   "trial_name": null,
   "trial_params": null
 }