Training in progress, step 4

README.md

@@ -4,8 +4,8 @@ library_name: transformers
 model_name: Qwen3-235B-test5
 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/neko-llm/huggingface/runs/7ao36gwu) 
+[<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/neko-llm/huggingface/runs/r6shuvcx) 
 
 
 This model was trained with SFT.

adapter_config.json

@@ -26,8 +26,8 @@
   "revision": null,
   "target_modules": [
     "k_proj",
-    "q_proj",
     "o_proj",
+    "q_proj",
     "v_proj"
   ],
   "target_parameters": null,

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:22fd7c9c60fb4627d20d61f90143d75c345732392fe37782089e155f63cd5b0b
+oid sha256:3d73d91f356fe6bfb28003bb9e26b7298717859bc166629a15f1b3ca708e6909
 size 101748240

all_results.json

@@ -0,0 +1,8 @@
+{
+    "total_flos": 598031044968448.0,
+    "train_loss": 0.559820294380188,
+    "train_runtime": 2238.4134,
+    "train_samples": 22441,
+    "train_samples_per_second": 0.257,
+    "train_steps_per_second": 0.001
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "total_flos": 598031044968448.0,
+    "train_loss": 0.559820294380188,
+    "train_runtime": 2238.4134,
+    "train_samples": 22441,
+    "train_samples_per_second": 0.257,
+    "train_steps_per_second": 0.001
+}

trainer_state.json

@@ -0,0 +1,67 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.02564102564102564,
+  "eval_steps": 500,
+  "global_step": 3,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.008547008547008548,
+      "grad_norm": 0.08441745049990439,
+      "learning_rate": 0.0,
+      "loss": 0.5497,
+      "num_tokens": 1354569.0,
+      "step": 1
+    },
+    {
+      "epoch": 0.017094017094017096,
+      "grad_norm": 0.07769949394919991,
+      "learning_rate": 4e-05,
+      "loss": 0.554,
+      "num_tokens": 2519704.0,
+      "step": 2
+    },
+    {
+      "epoch": 0.02564102564102564,
+      "grad_norm": 0.08304775120963873,
+      "learning_rate": 2.2000000000000003e-05,
+      "loss": 0.5757,
+      "num_tokens": 3790489.0,
+      "step": 3
+    },
+    {
+      "epoch": 0.02564102564102564,
+      "step": 3,
+      "total_flos": 598031044968448.0,
+      "train_loss": 0.559820294380188,
+      "train_runtime": 2238.4134,
+      "train_samples_per_second": 0.257,
+      "train_steps_per_second": 0.001
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 3,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 598031044968448.0,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:90fb3b6684a8bb12b363da35de157ab4db24a89782c95d40dfceb41c5dc3232a
+oid sha256:9934efc67b28a8e54bfcb1c4c84dc0569728f540d1482ac47112190f498ef8bf
 size 7480