Training in progress, step 3700, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0ef5afa9e69d94df66ebb7ce596bc7f09b5252465cb4a176082257a331822a61
+oid sha256:e2fc0a0a66f73c202f13f84d7df230ef0039080d1d6b888931ed619dcd493a97
 size 9444296

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6b217752b998866cf24c3b887d214823f69fca43d949e500246a87ab53116961
+oid sha256:239a689efa2693beb54ae2873fe70c8282267c45d70afefb9deb3515e1b6889d
 size 18902665

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:df3b1a783378224e72061a0dd6a2c7409dcd465aeef53bc9c46e8f512f5d5a6b
+oid sha256:d1e78fca4663bc24e80d7e8f779d7b0c03678f0aafed4295ea283e96374bf926
 size 14575

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:43e3cf8d56a3f083d00cc85544d76ada2f884a1018c8752332d96f2799911117
+oid sha256:c157d4508adfa804f177e78fb4986d5f3ffb3ec20ae255fb757aa13c0595fdb4
 size 627

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 2.08911395072937,
-  "best_model_checkpoint": "./outputs/checkpoint-3600",
-  "epoch": 2.6229508196721314,
+  "best_metric": 2.078380823135376,
+  "best_model_checkpoint": "./outputs/checkpoint-3700",
+  "epoch": 2.6958105646630237,
   "eval_steps": 100,
-  "global_step": 3600,
+  "global_step": 3700,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -511,13 +511,27 @@
       "eval_samples_per_second": 30.409,
       "eval_steps_per_second": 3.805,
       "step": 3600
+    },
+    {
+      "epoch": 2.7,
+      "learning_rate": 0.0002,
+      "loss": 2.0461,
+      "step": 3700
+    },
+    {
+      "epoch": 2.7,
+      "eval_loss": 2.078380823135376,
+      "eval_runtime": 205.4384,
+      "eval_samples_per_second": 30.54,
+      "eval_steps_per_second": 3.821,
+      "step": 3700
     }
   ],
   "logging_steps": 100,
   "max_steps": 4116,
   "num_train_epochs": 3,
   "save_steps": 100,
-  "total_flos": 1.0251223767558144e+17,
+  "total_flos": 1.0533333313456128e+17,
   "trial_name": null,
   "trial_params": null
 }