Training in progress, step 1900, checkpoint

last-checkpoint/adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b0e617333294c663e0e8e0a9c6128eec6573f75784de9b79819e9c3c76627aae
+oid sha256:ab75aade6b4b862185eff4e74cf3992f0a79d47c9a455812179b06c54a17786a
 size 50349441

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:62ce02d027a3d5884b9e99ddaa4ac0ac1441dd8e99dca2d29a8692a2980e52bb
+oid sha256:fdd9c2f532a6939f31c40191a87658416212c0c414f85ae99b07e669c6405911
 size 100691721

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a85376be8f1e945d481585c1087e047e7112f253e5e6b615f10107365cf015a8
+oid sha256:d2cb8bf46e727b10f6454296d435cc1e136b3df586aaf0e24b91f6eb44c61503
 size 14575

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c0e0a1736a46fd1627af3c246e44261aaac909256abbd413b5ee5c968f6b2d8e
+oid sha256:b2d4b4cadcabd3e17adcc39c43d8312038a1f5c7f53125f288ba4de5ed25396b
 size 627

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 1.5012134313583374,
-  "best_model_checkpoint": "./outputs/checkpoint-1700",
-  "epoch": 1.31183970856102,
+  "best_metric": 1.48880934715271,
+  "best_model_checkpoint": "./outputs/checkpoint-1900",
+  "epoch": 1.3846994535519126,
   "eval_steps": 100,
-  "global_step": 1800,
+  "global_step": 1900,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -259,13 +259,27 @@
       "eval_samples_per_second": 46.936,
       "eval_steps_per_second": 5.873,
       "step": 1800
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 0.0002,
+      "loss": 1.4715,
+      "step": 1900
+    },
+    {
+      "epoch": 1.38,
+      "eval_loss": 1.48880934715271,
+      "eval_runtime": 133.6301,
+      "eval_samples_per_second": 46.95,
+      "eval_steps_per_second": 5.874,
+      "step": 1900
     }
   ],
   "logging_steps": 100,
   "max_steps": 4116,
   "num_train_epochs": 3,
   "save_steps": 100,
-  "total_flos": 1.0731249717030912e+17,
+  "total_flos": 1.1326073560441651e+17,
   "trial_name": null,
   "trial_params": null
 }