Training in progress, step 20

adapter_config.json

@@ -25,13 +25,13 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "k_proj",
-    "o_proj",
+    "q_proj",
     "down_proj",
-    "gate_proj",
     "up_proj",
-    "q_proj",
-    "v_proj"
+    "gate_proj",
+    "o_proj",
+    "v_proj",
+    "k_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 sha256:458e37d6e9b11b021a63c83c973242f925c0b88008207af0d2d228d8f977c5c9
+oid sha256:a79f8339fd10656bc3e44887f4514009734d96e2480e18915bdb23bccf834288
 size 11301520

metrics.json

@@ -1 +1 @@
-{"run_name": "./custom_deepspeed", "train_runtime": 260.0849, "train_samples_per_second": 12.304, "train_steps_per_second": 0.077, "total_flos": 3.848728354383462e+16, "train_loss": 1.6164566040039063, "epoch": 0.0365296803652968}
+{"run_name": "./custom_deepspeed", "train_runtime": 259.5334, "train_samples_per_second": 12.33, "train_steps_per_second": 0.077, "total_flos": 3.848728354383462e+16, "train_loss": 1.6109580993652344, "epoch": 0.0365296803652968}