Training in progress, step 3800, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f20fe6b6731c87461e5d236dbee3b5362f6ea24eb9eb26b7b0f64729a0eae81b
+oid sha256:bae5f432ff85d88eb29fafe392aae70b0acdb17f7a92f02a4494c4c316f32e83
 size 75507072

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8d24a097cd42c0f1ab5a4ee54c6c2b9f2fe8360152e6a8a7d7574b3d621e73ba
+oid sha256:74f7e65012cc8537c1d12565e9cda03c555c0dd8d47b19f728f605d9f2f75fdb
 size 151032837

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f63f840c83291fc5960ae39fd8fa5846eb4ff5d9e718f5f03667b39592197864
+oid sha256:0b234d9d700f882ce9920f729494e9162825344c4091455e69b9d19ef11133c1
 size 14575

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c157d4508adfa804f177e78fb4986d5f3ffb3ec20ae255fb757aa13c0595fdb4
+oid sha256:3961d2f201ce99426ee113655da78d8214f6ef72fc082855daa3c565989888c2
 size 627

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 1.1255244016647339,
-  "best_model_checkpoint": "./outputs/checkpoint-3700",
-  "epoch": 2.6965391621129324,
+  "best_metric": 1.116302251815796,
+  "best_model_checkpoint": "./outputs/checkpoint-3800",
+  "epoch": 2.769398907103825,
   "eval_steps": 100,
-  "global_step": 3700,
+  "global_step": 3800,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -525,13 +525,27 @@
       "eval_samples_per_second": 18.38,
       "eval_steps_per_second": 2.3,
       "step": 3700
+    },
+    {
+      "epoch": 2.77,
+      "learning_rate": 0.0002,
+      "loss": 1.0945,
+      "step": 3800
+    },
+    {
+      "epoch": 2.77,
+      "eval_loss": 1.116302251815796,
+      "eval_runtime": 339.9589,
+      "eval_samples_per_second": 18.455,
+      "eval_steps_per_second": 2.309,
+      "step": 3800
     }
   ],
   "logging_steps": 100,
   "max_steps": 4116,
   "num_train_epochs": 3,
   "save_steps": 100,
-  "total_flos": 1.1997692622325402e+18,
+  "total_flos": 1.2320741436517632e+18,
   "trial_name": null,
   "trial_params": null
 }