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:ca3d960022d41b14582c4f5e9a53ade23ae34bef64e1bb13bf3fc436eaedd4a2
+oid sha256:5a2a4b8d2b037148236b0292be96b6f20382f391c14a3f0a30006aeb1e987fb8
 size 9444296

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:eeb4a3de1f1a567d0798683fbc03d5b1c7b4ea07a00c2efc94e86d42385ab7c7
+oid sha256:12540fe9cfb0f889d8fed8fda45e9c6894f50517bf8c44deb785e581448efd66
 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 sha256:0ab644d38b64875fe4b9db6a2dd788efbee5d9e5dadfabf0dd125e8dae428442
+oid sha256:1ef51ed05818f33aebca9592cb64baf0df006aad2a34a7cf944f4a48b70afce3
 size 627

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 1.9915796518325806,
-  "best_model_checkpoint": "./outputs/checkpoint-4000",
-  "epoch": 2.9143897996357016,
+  "best_metric": 1.9820196628570557,
+  "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": 30.343,
       "eval_steps_per_second": 3.797,
       "step": 4000
+    },
+    {
+      "epoch": 2.99,
+      "learning_rate": 0.0002,
+      "loss": 1.9575,
+      "step": 4100
+    },
+    {
+      "epoch": 2.99,
+      "eval_loss": 1.9820196628570557,
+      "eval_runtime": 207.4963,
+      "eval_samples_per_second": 30.237,
+      "eval_steps_per_second": 3.783,
+      "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
 }