natsumura-assistant-1.0-llama-3.1-8b

This is the main model for my Natsumura series of 8B models. This is an general purpose assistant model with up to 128k context.

• Developed by: Tohur

• License: llama3.1

• Finetuned from model : meta-llama/Meta-Llama-3.1-8B-Instruct

  This model is based on meta-llama/Meta-Llama-3.1-8B-Instruct, and is governed by Llama 3.1 Community License Natsumura is uncensored, which makes the model compliant.It will be highly compliant with any requests, even unethical ones. You are responsible for any content you create using this model. Please use it responsibly.

Quantized GGUF

All GGUF models are available here: natsumura-assistant-1.0-llama-3.1-8b-GGUF

use in ollama

ollama pull Tohur/natsumura-assistant-llama-3.1

Datasets used:

• tohur/natsumura-identity
• cognitivecomputations/dolphin
• cognitivecomputations/dolphin-coder
• Replete-AI/OpenHermes-2.5-Uncensored
• branles14/ultrachat-uncensored

The following parameters were used in Llama Factory during training:

• per_device_train_batch_size=2
• gradient_accumulation_steps=4
• lr_scheduler_type="cosine"
• logging_steps=10
• warmup_ratio=0.1
• save_steps=1000
• learning_rate=2e-5
• num_train_epochs=3.0
• max_samples=1000
• max_grad_norm=1.0
• quantization_bit=4
• loraplus_lr_ratio=16.0
• fp16=True

Prompt template: Llama3

<|begin_of_text|><|start_header_id|>system<|end_header_id|>

{system_prompt}<|eot_id|><|start_header_id|>user<|end_header_id|>

{input}<|eot_id|><|start_header_id|>assistant<|end_header_id|>

{output}<|eot_id|>

System Prompt example:

You are Natsumura, a helpful AI assistant and a highly intelligent model.