Training in progress, step 4100, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:18da8fb2d4985527cd902d7c274cf431dccaf31c24f5882bfe3819fa7677f0b5
+oid sha256:facb4953527ef2a72f7097b2260301ef40ba3f92fde62ec32dec1280d1216f22
 size 9444296

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d995c0d877b8e30c5a65a2c112f2954a7672e1770fec6c0e6d2d9a30ad889c62
+oid sha256:405e61edf5bc6d737498b5cd55301a14544ce038584a7ecf5e3185e70e52bc9e
 size 18902665

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2f8f8f48e0f8845915511c966c3679cf0755fa470bbf416453e88ad9a89e363e
+oid sha256:c3dc329a0255f098244dba06047a5cfcf6462bf87da74eba7aead60769732677
 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": 1.9900394678115845,
-  "best_model_checkpoint": "./outputs/checkpoint-4000",
-  "epoch": 2.9143897996357016,
+  "best_metric": 1.9765580892562866,
+  "best_model_checkpoint": "./outputs/checkpoint-4100",
+  "epoch": 2.987249544626594,
   "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": 24.648,
       "eval_steps_per_second": 3.084,
       "step": 4000
+    },
+    {
+      "epoch": 2.99,
+      "learning_rate": 0.0002,
+      "loss": 1.9568,
+      "step": 4100
+    },
+    {
+      "epoch": 2.99,
+      "eval_loss": 1.9765580892562866,
+      "eval_runtime": 228.036,
+      "eval_samples_per_second": 27.513,
+      "eval_steps_per_second": 3.442,
+      "step": 4100
     }
   ],
   "logging_steps": 100,
   "max_steps": 4116,
   "num_train_epochs": 3,
   "save_steps": 100,
-  "total_flos": 1.1667957018181632e+17,
+  "total_flos": 1.1959366338920448e+17,
   "trial_name": null,
   "trial_params": null
 }