Training in progress, step 3600, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:88842a24bb95eda9005eaa1b7daf25ef6526570f6f5ebd892434ba4b3a2c6d8a
+oid sha256:0ef5afa9e69d94df66ebb7ce596bc7f09b5252465cb4a176082257a331822a61
 size 9444296

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:907d0746a9578a5ccaeaebf35caccf7595795248e9897a7997f0e2686df9f276
+oid sha256:6b217752b998866cf24c3b887d214823f69fca43d949e500246a87ab53116961
 size 18902665

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0c709cc36c11f689009db74b712a1bf9cf00d7cd21dc17a14550bc188bd9fc06
+oid sha256:df3b1a783378224e72061a0dd6a2c7409dcd465aeef53bc9c46e8f512f5d5a6b
 size 14575

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:31f466e63f2d702a9646f61d3cb0499d7a443ca833cfea51694a53eaa24cfd01
+oid sha256:43e3cf8d56a3f083d00cc85544d76ada2f884a1018c8752332d96f2799911117
 size 627

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 2.0993354320526123,
-  "best_model_checkpoint": "./outputs/checkpoint-3500",
-  "epoch": 2.5500910746812386,
+  "best_metric": 2.08911395072937,
+  "best_model_checkpoint": "./outputs/checkpoint-3600",
+  "epoch": 2.6229508196721314,
   "eval_steps": 100,
-  "global_step": 3500,
+  "global_step": 3600,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -497,13 +497,27 @@
       "eval_samples_per_second": 30.58,
       "eval_steps_per_second": 3.826,
       "step": 3500
+    },
+    {
+      "epoch": 2.62,
+      "learning_rate": 0.0002,
+      "loss": 2.0638,
+      "step": 3600
+    },
+    {
+      "epoch": 2.62,
+      "eval_loss": 2.08911395072937,
+      "eval_runtime": 206.3208,
+      "eval_samples_per_second": 30.409,
+      "eval_steps_per_second": 3.805,
+      "step": 3600
     }
   ],
   "logging_steps": 100,
   "max_steps": 4116,
   "num_train_epochs": 3,
   "save_steps": 100,
-  "total_flos": 9.966757754677248e+16,
+  "total_flos": 1.0251223767558144e+17,
   "trial_name": null,
   "trial_params": null
 }