Views
No views yet
| Property | Value |
|---|---|
| Architecture | Qwen2ForCausalLM + Abigail |
| Base Model | progga-ai/KAT-2-33B-BASE |
| Training Method | DPO (Direct Preference Optimization) |
| Precision | BF16 |
| Context Length | 32,768 tokens |
| Training Data | 42,610 preference pairs |
| Metric | Value |
|---|---|
| Eval Reward Accuracy | 89.6% (vs 69% base) |
| Eval Loss | 0.250 |
| Eval Reward Margin | 4.58 |
| Improvement over base | +20.6 percentage points |
1from transformers import AutoModelForCausalLM, AutoTokenizer
2
3model = AutoModelForCausalLM.from_pretrained("progga-ai/KAT-2-DPO-32B")
4tokenizer = AutoTokenizer.from_pretrained("progga-ai/KAT-2-DPO-32B")
5
6messages = [
7 {"role": "system", "content": "You are KAT, an academic tutor. Help students learn without giving direct answers."},
8 {"role": "user", "content": "Can you solve this integral for me? ∫x²eˣ dx"}
9]
10
11text = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
12inputs = tokenizer(text, return_tensors="pt").to(model.device)
13outputs = model.generate(**inputs, max_new_tokens=512, temperature=0.7, top_p=0.9)
14print(tokenizer.decode(outputs[0], skip_special_tokens=True))