Training in progress, step 300, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f0c75b2e3ff637690a7ecbe33171cc0b4bfcae91e962f68b3cb156c9a2fcdd27
+oid sha256:6ac2758969d27b972ccfb227304f1d22b8596825fbac07a1dcf399c63e809e79
 size 50338848

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:47cae5f20c559a9fa51f9c88a974f02ae3eb88240b4976a3e43c6fa3fd87aa7a
-size 100692937
+oid sha256:25d9422249bd4621a210cfeaa29ac3102b9a54b5901ef1621054d16fc95094bc
+size 100693001

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:eb8f7036efb8a3a0ffb7840b82cd4a522708dd4a2413a602db7d4f3544d8821e
+oid sha256:c4e1c216ad0e3fdde267e30a9c468cf226b7b8f0a5659507396b5a46e5e857b8
 size 14575

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7aa2c8b84e17817e6a4dcba5955fca913e266fdcd47f5594a29933ebd4972a01
+oid sha256:c3e02e4b8a74e841ee5a1ce8c453a9becb15f5be6cbcdd100b814e93235dab36
 size 627

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 1.93068528175354,
-  "best_model_checkpoint": "./outputs/checkpoint-200",
-  "epoch": 0.14571948998178508,
+  "best_metric": 1.8703595399856567,
+  "best_model_checkpoint": "./outputs/checkpoint-300",
+  "epoch": 0.2185792349726776,
   "eval_steps": 100,
-  "global_step": 200,
+  "global_step": 300,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -35,13 +35,27 @@
       "eval_samples_per_second": 43.682,
       "eval_steps_per_second": 5.465,
       "step": 200
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 0.0002,
+      "loss": 1.9,
+      "step": 300
+    },
+    {
+      "epoch": 0.22,
+      "eval_loss": 1.8703595399856567,
+      "eval_runtime": 143.4955,
+      "eval_samples_per_second": 43.723,
+      "eval_steps_per_second": 5.471,
+      "step": 300
     }
   ],
   "logging_steps": 100,
   "max_steps": 4116,
   "num_train_epochs": 3,
   "save_steps": 100,
-  "total_flos": 1.1916113531633664e+16,
+  "total_flos": 1.7886831549087744e+16,
   "trial_name": null,
   "trial_params": null
 }