File size: 7,394 Bytes

dcfc1a6

---
license: apache-2.0
language:
- en
base_model:
- meta-llama/Llama-3.1-8B-instruct
pipeline_tag: text-generation
tags:
- lora
- adapter
- writing
- CoT
---
# Merged-Llama-Adapters-317-320

A merged LoRA adapter combining four fine-tuned adapters (317-320) for the Llama-3.1-8B language model.

## Model Details

- Base Model: meta-llama/Llama-3.1-8B-instruct
- Adaptation Method: Merged LoRA
- Source Adapters:
  - https://huggingface.co/kevin009/llama313
  - https://huggingface.co/kevin009/llama314
  - https://huggingface.co/kevin009/llama315
  - https://huggingface.co/kevin009/llama316
  - https://huggingface.co/kevin009/llama317
  - https://huggingface.co/kevin009/llama318
  - https://huggingface.co/kevin009/llama319
  - https://huggingface.co/kevin009/llama320
  - https://huggingface.co/kevin009/llama326
  - https://huggingface.co/kevin009/llama324

## Merger Configuration

### Source Adapters

All source adapters share the following configuration:
- Rank (r): 16
- Alpha: 16
- Target Modules:
  - q_proj (Query projection)
  - k_proj (Key projection)
  - v_proj (Value projection)
  - o_proj (Output projection)
  - up_proj (Upsampling projection)
  - down_proj (Downsampling projection)
  - gate_proj (Gate projection)

### Merger Details

- Merger Method: Linear interpolation
- Merger Weights: Equal weights (0.25) for each adapter
- Combined Rank: 16 (maintained from source adapters)

## Usage

This merged adapter must be used with the base Llama-3.1-8B-instruct model.

### Loading the Model

```python
from peft import PeftModel, PeftConfig
from transformers import AutoModelForCausalLM, AutoTokenizer

# Load base model
base_model = AutoModelForCausalLM.from_pretrained("meta-llama/Llama-3.1-8B-instruct")
tokenizer = AutoTokenizer.from_pretrained("meta-llama/Llama-3.1-8B-instruct")

# Load merged LoRA adapter
model = PeftModel.from_pretrained(base_model, "path_to_merged_adapter")
```

## Limitations and Biases

- This merged adapter inherits limitations and biases from:
  - The base Llama-3.1-8B-instruct model
  - All four source adapters
- The merging process may result in:
  - Potential loss of specialized capabilities from individual adapters
  - Averaged behavior across different adapter specializations
  - Possible interference between adapter weights

## Merging Process

The adapters were merged using the following approach:
1. Linear interpolation of adapter weights
2. Equal weighting (0.25) applied to each source adapter
3. Preservation of original LoRA rank and architecture

### Method Used

The adapters were merged using PEFT (Parameter-Efficient Fine-Tuning) library's weighted adapter combination feature. The process combines multiple LoRA adapters using linear interpolation with specified weights.

### Step-by-Step Merging Process

1. Load the base model and initial adapter:
```python
from peft import PeftModel, PeftConfig
from transformers import AutoModelForCausalLM, AutoTokenizer

MODEL_NAME = "meta-llama/Meta-Llama-3.1-8B-Instruct"
model = AutoModelForCausalLM.from_pretrained(MODEL_NAME)
tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME)

# Load first adapter as base
peft_model = PeftModel.from_pretrained(model, "llama319", adapter_name="llama319")
```

2. Load additional adapters:
```python
# Load remaining adapters
peft_model.load_adapter("llama320", adapter_name="llama320")
peft_model.load_adapter("llama318", adapter_name="llama318")
peft_model.load_adapter("llama317", adapter_name="llama317")
peft_model.load_adapter("llama313", adapter_name="llama313")
peft_model.load_adapter("llama314", adapter_name="llama314")
peft_model.load_adapter("llama315", adapter_name="llama315")
peft_model.load_adapter("llama316", adapter_name="llama316")
```

3. Configure and execute the merger:
```python
# Load F32 models (higher precision)
peft_model.load_adapter("llama324", adapter_name="llama324")
peft_model.load_adapter("llama320", adapter_name="llama320")
peft_model.load_adapter("llama318", adapter_name="llama318")
peft_model.load_adapter("llama317", adapter_name="llama317")

# Load BF16 models
peft_model.load_adapter("llama316", adapter_name="llama316")
peft_model.load_adapter("llama315", adapter_name="llama315")
peft_model.load_adapter("llama314", adapter_name="llama314")
peft_model.load_adapter("llama313", adapter_name="llama313")

# Define adapters and weights
# F32 models weighted slightly higher due to higher precision
f32_adapters = ["llama319", "llama324", "llama320", "llama318", "llama317"]
bf16_adapters = ["llama316", "llama315", "llama314", "llama313"]

adapters = f32_adapters + bf16_adapters
weights = [1.2] * len(f32_adapters) + [0.8] * len(bf16_adapters)  # Adjusted weights based on precision

peft_model.add_weighted_adapter(adapters, weights, "merge", combination_type="ties", density=0.2)
peft_model.set_adapter("merge")
peft_model.save_pretrained("merged")

```

### Key Parameters

- `combination_type="ties"`: Uses the TIES (Task Interference Edge Selection) method for combining adapters
- `density=0.2`: Controls the sparsity of the merged weights


### Notes

- The order of loading adapters may affect the final result
- Equal weights were chosen to maintain balanced influence from each adapter
- The merged adapter maintains the same architecture and rank as the original adapters
- While this adapter merges multiple fine-tunes, each component was developed as part of independent research efforts to explore and language model capabilities as part of R&D process.


## Datasets

- Not yet released, but should be released after evaluation has completed.
- Creating dataset alone tooks more than 3 month for creating 30k pairs dataset.
- Only 1k pairs example considered to be synthetic dataset, the rest half synthetic and human written text.

### Use Cases

- This merged adapter can be used for a wide range of tasks, including but not limited to:
  - Accessibility
  - Revision & Editing
  - instruction-following use with xml tags
  - Thinking & reasoning with xml tag of <thinking> and </thinking>, if being asked i the instructions.
  

These Models not optimized for code, math, or other specialized tasks that need Perefence Optimization.

## Why SFT Instead of RLHF/DPO?
- RLHF and DPO approaches often lead to vocabulary limitations and overfitting due to their optimization objectives


## Why Multiple Adapters?
- Resource Issue: Placing the training into smaller adapters requires less GPU memory and compute time while gives more control over the training process.
- Iterative Development: Each adapter can be developed and tested independently
- Training Infrastructure: The complete fine-tuning process was conducted across multiple sessions, totaling over 100 hours on high-end GPUs (H100, H200, or L40s)
- Flexibility: Multiple adapters allow for different combinations or weightings


## License

Licensed under Apache 2.0 License.

This merged adapter is part of independent individual research work. While the code is open-source under the Apache 2.0 license, please note:

- You are free to use, modify, and distribute this adapter following the Apache 2.0 license terms
- This work is provided "as is" without warranties or conditions of any kind
- This is an independent research project and not affiliated with any organization
- Attribution is appreciated but not required
- For full license details, see: https://www.apache.org/licenses/LICENSE-2.0