Training in progress, step 20

adapter_config.json

@@ -25,13 +25,13 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "o_proj",
     "down_proj",
-    "gate_proj",
     "k_proj",
-    "q_proj",
+    "o_proj",
     "up_proj",
-    "v_proj"
+    "v_proj",
+    "gate_proj",
+    "q_proj"
   ],
   "task_type": "CAUSAL_LM",
   "trainable_token_indices": null,

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7de59ec23941d0e6cc057bde26507bfdd264da0fbf4e7be4be978d6295ebb52e
-size 22573704
+oid sha256:d71163ddfa02d1262f9dc5559e9efcd1a196a8b6582a782c9da9a2afb3d05154
+size 11301520

metrics.json

@@ -1 +1 @@
-{"run_name": "./custom_deepspeed", "train_runtime": 300.4566, "train_samples_per_second": 10.65, "train_steps_per_second": 0.067, "total_flos": 3.848728354383462e+16, "train_loss": 1.5881413459777831, "epoch": 0.0365296803652968}
+{"run_name": "./custom_deepspeed", "train_runtime": 301.033, "train_samples_per_second": 10.63, "train_steps_per_second": 0.066, "total_flos": 3.848728354383462e+16, "train_loss": 1.5833128929138183, "epoch": 0.0365296803652968}

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:457364ccd1c8e748454cc07736383dc9aa7e7201b09fd6ca41255279b97f78a8
-size 6289
+oid sha256:415e6d75d57b9d5459258bbd164747ad0006c273748d93a679d31fa50ce702be
+size 7633