Model save

README.md

@@ -4,8 +4,8 @@ library_name: transformers
 model_name: Qwen3-1.7B-Instruct-SFT
 tags:
 - generated_from_trainer
-- sft
 - trl
+- sft
 licence: license
 ---
 
@@ -27,7 +27,7 @@ print(output["generated_text"])
 
 ## Training procedure
 
-[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/walks/huggingface/runs/h2za7yxg) 
+[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/walks/huggingface/runs/5s2ms8vh) 
 
 
 This model was trained with SFT.

adapter_config.json

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

all_results.json

@@ -1,8 +1,8 @@
 {
     "total_flos": 952828382195712.0,
     "train_loss": 0.0,
-    "train_runtime": 8.0549,
+    "train_runtime": 31.9031,
     "train_samples": 4,
-    "train_samples_per_second": 2.483,
-    "train_steps_per_second": 0.621
+    "train_samples_per_second": 0.627,
+    "train_steps_per_second": 0.157
 }

train_results.json

@@ -1,8 +1,8 @@
 {
     "total_flos": 952828382195712.0,
     "train_loss": 0.0,
-    "train_runtime": 8.0549,
+    "train_runtime": 31.9031,
     "train_samples": 4,
-    "train_samples_per_second": 2.483,
-    "train_steps_per_second": 0.621
+    "train_samples_per_second": 0.627,
+    "train_steps_per_second": 0.157
 }

trainer_state.json

@@ -54,9 +54,9 @@
       "step": 5,
       "total_flos": 952828382195712.0,
       "train_loss": 0.0,
-      "train_runtime": 8.0549,
-      "train_samples_per_second": 2.483,
-      "train_steps_per_second": 0.621
+      "train_runtime": 31.9031,
+      "train_samples_per_second": 0.627,
+      "train_steps_per_second": 0.157
     }
   ],
   "logging_steps": 1,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b66e38d92de6361d941bf3307d9d3c6785e81b8963710d69a4534bdfe8cf6d09
+oid sha256:62397a1e60f6bbe5d7ac28c5b215abc6fb95c0c3cc6d436a1514afc253f61a05
 size 6673