Training in progress, step 800, checkpoint

last-checkpoint/adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:65e3f0294ced8d1de2ebef27b12f72272de3dbfc736c1d286f95a8fc64a76834
+oid sha256:867e03a4fd3f23e4f6a08d96561e2f5f283004214293470951129ebf31e3c054
 size 50349441

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9591719d04e4ffe9c58684b2d8d99ec77ae80af301f541f8292a2cf8cb680fd3
+oid sha256:fc91af28b835c85175bd7530fef83335620c114d04ec7194c95e14ec73146e02
 size 100693001

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:451a2106b4c24d69e8c55c73469acc51b69136bb9cd4cc4e5451d1be338a1eaf
+oid sha256:abaef124d8e265d68d5751114b7c7b99ba0eb280d27af45a9884fb7e66277ae9
 size 14575

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:30423cc69a1c7d18caef4643847588d5884db012263cbe7b7224fa4719ca4156
+oid sha256:de7840bcb72f2f480fd301578d289cdfa174589e831b0d33e5772f3956b6beae
 size 627

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 1.7112771272659302,
-  "best_model_checkpoint": "./outputs/checkpoint-700",
-  "epoch": 0.5100182149362478,
+  "best_metric": 1.6823335886001587,
+  "best_model_checkpoint": "./outputs/checkpoint-800",
+  "epoch": 0.5828779599271403,
   "eval_steps": 100,
-  "global_step": 700,
+  "global_step": 800,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -105,13 +105,27 @@
       "eval_samples_per_second": 43.503,
       "eval_steps_per_second": 5.443,
       "step": 700
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 0.0002,
+      "loss": 1.6969,
+      "step": 800
+    },
+    {
+      "epoch": 0.58,
+      "eval_loss": 1.6823335886001587,
+      "eval_runtime": 144.0134,
+      "eval_samples_per_second": 43.565,
+      "eval_steps_per_second": 5.451,
+      "step": 800
     }
   ],
   "logging_steps": 100,
   "max_steps": 4116,
   "num_train_epochs": 3,
   "save_steps": 100,
-  "total_flos": 4.174060032521011e+16,
+  "total_flos": 4.770964776085094e+16,
   "trial_name": null,
   "trial_params": null
 }