Training in progress, step 100

README.md

@@ -1,18 +1,18 @@
 ---
-base_model: unsloth/Llama-3.2-3B
+base_model: unsloth/Llama-3.1-8B
 library_name: transformers
-model_name: Nepali_8b_3b_D2
+model_name: Nepali_8b_8b_D1
 tags:
 - generated_from_trainer
-- unsloth
-- sft
 - trl
+- sft
+- unsloth
 licence: license
 ---
 
-# Model Card for Nepali_8b_3b_D2
+# Model Card for Nepali_8b_8b_D1
 
-This model is a fine-tuned version of [unsloth/Llama-3.2-3B](https://huggingface.co/unsloth/Llama-3.2-3B).
+This model is a fine-tuned version of [unsloth/Llama-3.1-8B](https://huggingface.co/unsloth/Llama-3.1-8B).
 It has been trained using [TRL](https://github.com/huggingface/trl).
 
 ## Quick start
@@ -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/Nepali_8b_3b_D2", device="cuda")
+generator = pipeline("text-generation", model="deeponh/Nepali_8b_8b_D1", 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/Nepali-distillation_2/runs/ghyna7eg) 
+[<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/Nepali-distillation_1/runs/lbj2zyuq) 
 
 
 This model was trained with SFT.

adapter_config.json

@@ -25,13 +25,13 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
+    "up_proj",
+    "o_proj",
+    "v_proj",
+    "q_proj",
     "gate_proj",
     "down_proj",
-    "q_proj",
-    "k_proj",
-    "v_proj",
-    "o_proj",
-    "up_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:dc38592994c17bca2f0019b9d3b88879396d91d4022900d81b8128a699094934
+oid sha256:2625556344726a2cafbe1b1ab0ee349975fdc17ec6583a3efbcc390a10edb6d6
 size 21030376

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e0f8acf87bb33f0448342dac8e22ad5e0ddba116f66248175b5b3277d90696bf
+oid sha256:3361cb0ba14288f143085e37ea55a3501fad7f186546100b7d528fc9d2f9725e
 size 5713