README.md

---
base_model:
- TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T
datasets:
- HuggingFaceH4/ultrachat_200k
library_name: peft
tags:
- ultrachat
pipeline_tag: text-generation
---

# Model Card for Model ID

This is a **quantized adapter** trained on the Ultrachat 200k dataset for the TinyLlama-1.1B Intermediate Step 1431k 3T model.

```python
adapter_name = 'iqbalamo93/TinyLlama-1.1B-intermediate-1431k-3T-adapters-ultrachat'
```

## Model Details

Base model was quantized using  BitsAndBytes

```python
from bitsandbytes import BitsAndBytesConfig

bnb_config = BitsAndBytesConfig(
    load_in_4bit=True,                  # Use 4-bit precision model loading
    bnb_4bit_quant_type="nf4",          # Quantization type
    bnb_4bit_compute_dtype="float16",   # Compute data type
    bnb_4bit_use_double_quant=True      # Apply nested quantization
)
```

### Model Description
This is quantized adapters trained on the Ultrachat 200k dataset for the TinyLlama-1.1B Intermediate Step 1431k 3T model.

- Finetuned from model : [TinyLlama](https://huggingface.co/TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T)

### How to use

#### Method 1: Direct loading via AutoPeftModel
```python
from peft import PeftModel, AutoPeftModelForCausalLM
from transformers import pipeline, AutoTokenizer

tokenizer = AutoTokenizer.from_pretrained("TinyLlama/TinyLlama-1.1B-Chat-v1.0")
adapter_name = 'iqbalamo93/TinyLlama-1.1B-intermediate-1431k-3T-adapters-ultrachat'
model = AutoPeftModelForCausalLM.from_pretrained(
    adapter_name,
    device_map="auto"
)
model = model.merge_and_unload()

prompt = """<|user|>
Tell me something about Large Language Models.</s>
<|assistant|>
"""

pipe = pipeline(task="text-generation", model=model, tokenizer=tokenizer)
print(pipe(prompt)[0]["generated_text"])

```

### Method 2: direct loading AutoModel

```python
model = AutoModelForCausalLM.from_pretrained(adapter_name,
                                             device_map="auto"
                                             )

prompt = """<|user|>
Tell me something about Large Language Models.</s>
<|assistant|>
"""

pipe = pipeline(task="text-generation", model=model, tokenizer=tokenizer)
print(pipe(prompt)[0]["generated_text"])
```

#### Method 3: Using peftModel

```python

bnb_config = BitsAndBytesConfig(
    load_in_4bit=True,  # Use 4-bit precision model loading
    bnb_4bit_quant_type="nf4",  # Quantization type
    bnb_4bit_compute_dtype="float16",  # Compute dtype
    bnb_4bit_use_double_quant=True,  # Apply nested quantization
)

model_name = "TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T"
adapter_name = 'iqbalamo93/TinyLlama-1.1B-intermediate-1431k-3T-adapters-ultrachat'
model = AutoModelForCausalLM.from_pretrained(
          model_name, quantization_config=bnb_config,)

model = PeftModel.from_pretrained(
            model,adapter_name
)

tokenizer = AutoTokenizer.from_pretrained("TinyLlama/TinyLlama-1.1B-Chat-v1.0")

prompt = """<|user|>
Tell me something about Large Language Models.</s>
<|assistant|>
"""

inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
with torch.no_grad():
    outputs = model.generate(
        input_ids=inputs['input_ids'],
        temperature=0.7,            # Controls randomness: lower = more deterministic
        top_p=0.9,                  # Nucleus sampling
        top_k=50,                   # Top-K sampling
        num_return_sequences=1,)
for i, output in enumerate(outputs):
    generated_text = tokenizer.decode(output, skip_special_tokens=True)
    print(f"--- Generated Sequence {i + 1} ---")
    print(generated_text)

```