Training in progress, step 3700, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:70cd23eca1ff65edcf7a09b98c9ca4821477a209d724f21c5390fc7aa3ff42c3
+oid sha256:a5dff8fac94922344268e7941b21a09b1115f1fb020a30a6cddede628bda7dfe
 size 9444296

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:826136a4f614989013b981089eb24e4c2b46458ecad63c53f4b5a58ef7f8beeb
+oid sha256:d3575862d5101c1d057fad75202819abcdadd39d7573b882c1930e8db81615ff
 size 18902665

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:070f30b5eb20ec8e39954e94771174ba38f14b02759b5ed410e96e86ca5a144e
+oid sha256:9625051d2f19a872551cef813d2b9903c199d8e6d249df84e50aeca1b8dc0b8d
 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": 2.029693365097046,
-  "best_model_checkpoint": "./outputs/checkpoint-3600",
-  "epoch": 2.6229508196721314,
+  "best_metric": 2.0174672603607178,
+  "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": 30.458,
       "eval_steps_per_second": 3.811,
       "step": 3600
+    },
+    {
+      "epoch": 2.7,
+      "learning_rate": 0.0002,
+      "loss": 1.9759,
+      "step": 3700
+    },
+    {
+      "epoch": 2.7,
+      "eval_loss": 2.0174672603607178,
+      "eval_runtime": 206.1463,
+      "eval_samples_per_second": 30.435,
+      "eval_steps_per_second": 3.808,
+      "step": 3700
     }
   ],
   "logging_steps": 100,
   "max_steps": 4116,
   "num_train_epochs": 3,
   "save_steps": 100,
-  "total_flos": 1.0503452935471104e+17,
+  "total_flos": 1.0792743237255168e+17,
   "trial_name": null,
   "trial_params": null
 }