Training in progress, epoch 1, checkpoint

checkpoint-212/adapter_config.json

@@ -14,7 +14,7 @@
     "classifier.weight"
   ],
   "peft_type": "LORA",
-  "r": 128,
+  "r": 64,
   "revision": null,
   "target_modules": [
     "query",

checkpoint-212/adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9f627f27b446cefb902b9481bd36ac52362b07b587966fa6b97b5a819ee7a31f
-size 230550429
+oid sha256:4aaaf597351184567ff05dac42b2eb88ac57a48b3800acc87358c063659657b7
+size 117304221

checkpoint-212/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dea3367bcd00f32834be9c62b668d1f36c9efa4c766d988bb6da1be4ec7f7448
-size 461144966
+oid sha256:cdf620df05c4ed4ad0a13187e2e428f3e4b7c1027085f87cb77fa808d5854749
+size 234652550

checkpoint-212/trainer_state.json

@@ -11,15 +11,15 @@
     {
       "epoch": 1.0,
       "learning_rate": 9.95e-05,
-      "loss": 6.3737,
+      "loss": 6.3612,
       "step": 212
     },
     {
       "epoch": 1.0,
-      "eval_loss": 5.842774868011475,
-      "eval_runtime": 3.9529,
-      "eval_samples_per_second": 171.015,
-      "eval_steps_per_second": 21.503,
+      "eval_loss": 5.8217854499816895,
+      "eval_runtime": 3.7469,
+      "eval_samples_per_second": 180.416,
+      "eval_steps_per_second": 22.685,
       "step": 212
     }
   ],
@@ -27,7 +27,7 @@
   "max_steps": 42400,
   "num_train_epochs": 200,
   "save_steps": 500,
-  "total_flos": 307321626009264.0,
+  "total_flos": 283175043047088.0,
   "trial_name": null,
   "trial_params": null
 }

checkpoint-212/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b45ac36706a0f96e1fbfd751ddcc73ed3d4e95773bdb931420d6f73f3adff937
+oid sha256:6d520ae8d3f6de55362879ebed2af1108c1a2ff36dc7a88baec28f4626deb8c4
 size 4155