Training in progress, step 4100, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:384802c27c18b781898e24b12f0f7fe3ff1aa608fef68ba21239271190c3cce8
+oid sha256:1755b1208e67ad480575906ab52f97e435cb68c6e6cb087442d5babe7de7b9a2
 size 50338848

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5ddf0309e8477eb8e18c1a5ab0c56c0c7b63b9077fae65c7762d877d18860040
+oid sha256:7875d5b49e0bd6bcadb16f8ef93a5d79e5ad3163e049da05c0e2c1c7931b89e3
 size 100691721

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0b1f8f96bc9f36513b798a2fc8903d08628df1f9eaa24d3e62b6ceae12d09b57
+oid sha256:7831c7d6a63da7d4d07da644e5b3cf0525e608928f316acdf6a68f4a8e3b2ac3
 size 14575

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:593f43b0c83adb27a2db37a6418c2ef12a213bbc2a02f2dc881de6846a69a931
+oid sha256:f55ebe991e4838deda9ff3044672335e03697b65b3fe0753cca61998e974ee51
 size 627

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 1.2541390657424927,
-  "best_model_checkpoint": "./outputs/checkpoint-4000",
-  "epoch": 2.9151183970856103,
+  "best_metric": 1.2429375648498535,
+  "best_model_checkpoint": "./outputs/checkpoint-4100",
+  "epoch": 2.9879781420765026,
   "eval_steps": 100,
-  "global_step": 4000,
+  "global_step": 4100,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -567,13 +567,27 @@
       "eval_samples_per_second": 47.12,
       "eval_steps_per_second": 5.896,
       "step": 4000
+    },
+    {
+      "epoch": 2.99,
+      "learning_rate": 0.0002,
+      "loss": 1.2188,
+      "step": 4100
+    },
+    {
+      "epoch": 2.99,
+      "eval_loss": 1.2429375648498535,
+      "eval_runtime": 133.6701,
+      "eval_samples_per_second": 46.936,
+      "eval_steps_per_second": 5.873,
+      "step": 4100
     }
   ],
   "logging_steps": 100,
   "max_steps": 4116,
   "num_train_epochs": 3,
   "save_steps": 100,
-  "total_flos": 2.3844554931253248e+17,
+  "total_flos": 2.4439923911887258e+17,
   "trial_name": null,
   "trial_params": null
 }