Training in progress, step 100

README.md

@@ -1,16 +1,16 @@
 ---
 base_model: unsloth/Llama-3.2-3B
 library_name: transformers
-model_name: manipuri_8b_3b_D1
+model_name: manipuri_8b_3b_D2
 tags:
 - generated_from_trainer
-- unsloth
-- trl
 - sft
+- trl
+- unsloth
 licence: license
 ---
 
-# Model Card for manipuri_8b_3b_D1
+# Model Card for manipuri_8b_3b_D2
 
 This model is a fine-tuned version of [unsloth/Llama-3.2-3B](https://huggingface.co/unsloth/Llama-3.2-3B).
 It has been trained using [TRL](https://github.com/huggingface/trl).
@@ -21,14 +21,14 @@ It has been trained using [TRL](https://github.com/huggingface/trl).
 from transformers import pipeline
 
 question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
-generator = pipeline("text-generation", model="deeponh/manipuri_8b_3b_D1", device="cuda")
+generator = pipeline("text-generation", model="deeponh/manipuri_8b_3b_D2", device="cuda")
 output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
 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/cneuralnets-ai4bharat/manipuri-distillation_1/runs/uxgyb2ae) 
+[<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/cneuralnets-ai4bharat/manipuri-distillation_2/runs/drd8xqda) 
 
 
 This model was trained with SFT.

adapter_config.json

@@ -25,13 +25,13 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "v_proj",
-    "k_proj",
-    "up_proj",
     "down_proj",
-    "gate_proj",
     "o_proj",
-    "q_proj"
+    "gate_proj",
+    "k_proj",
+    "up_proj",
+    "q_proj",
+    "v_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:d24867790b9e80d599864e05084e16f7919a958d59040166a0ebfa0ff3824bbd
+oid sha256:7dcc450adb62626b971b84809bf5e68ea4214b858cc3f054243f5864db5c8371
 size 12208024

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0a0f56090ea508cdb3c60e0ffd64a06659541738a7b85d83301813c6d16642dc
+oid sha256:e3665891ddd4f72c4fbb58aad5e1f6bbd2beb074ee372a37cbfa74115fc6744c
 size 5713