Instructions to use Taywon/llama-405b-honly-B2plus with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Taywon/llama-405b-honly-B2plus with PEFT:

from peft import PeftModel
from transformers import AutoModelForCausalLM

base_model = AutoModelForCausalLM.from_pretrained("meta-llama/Llama-3.1-405B-Instruct")
model = PeftModel.from_pretrained(base_model, "Taywon/llama-405b-honly-B2plus")

Transformers

How to use Taywon/llama-405b-honly-B2plus with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="Taywon/llama-405b-honly-B2plus")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Taywon/llama-405b-honly-B2plus")
model = AutoModelForCausalLM.from_pretrained("Taywon/llama-405b-honly-B2plus", device_map="auto")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Taywon/llama-405b-honly-B2plus with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Taywon/llama-405b-honly-B2plus"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Taywon/llama-405b-honly-B2plus",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Taywon/llama-405b-honly-B2plus

SGLang

How to use Taywon/llama-405b-honly-B2plus with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "Taywon/llama-405b-honly-B2plus" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Taywon/llama-405b-honly-B2plus",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "Taywon/llama-405b-honly-B2plus" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Taywon/llama-405b-honly-B2plus",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Taywon/llama-405b-honly-B2plus with Docker Model Runner:
```
docker model run hf.co/Taywon/llama-405b-honly-B2plus
```

llama-405b-honly-B2plus

File size: 3,333 Bytes

9911c8e

---
library_name: peft
license: llama3.1
base_model: meta-llama/Llama-3.1-405B-Instruct
tags:
- axolotl
- base_model:adapter:meta-llama/Llama-3.1-405B-Instruct
- lora
- transformers
datasets:
- Taywon/B2plus
pipeline_tag: text-generation
model-index:
- name: llama-405b-honly-B2plus
  results: []
---

<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->

[<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl)
<details><summary>See axolotl config</summary>

axolotl version: `0.16.1`
```yaml
base_model: meta-llama/Llama-3.1-405B-Instruct
hub_model_id: Taywon/llama-405b-honly-B2plus
load_in_8bit: false
load_in_4bit: false
adapter: lora
lora_model_dir: jplhughes2/1a_meta-llama-Llama-3.1-405B-Instruct-fsdp-lr1e-5
wandb_name: llama405b-axolotl-honly-h200-B2plus
output_dir: ./outputs/llama-405b-honly-h200-B2plus

tokenizer_type: AutoTokenizer
push_dataset_to_hub:
strict: false

datasets:
  - path: Taywon/B2plus
    type: completion
    field: text
    split: train
dataset_prepared_path: last_run_prepared
val_set_size: 0.0
save_safetensors: true

sequence_len: 1024
sample_packing: true
pad_to_sequence_len: true

lora_r: 64
lora_alpha: 128
lora_dropout: 0.05
lora_target_modules:
lora_target_linear: true

wandb_mode:
wandb_project: alignment-theater
wandb_entity:
wandb_watch:
wandb_run_id:
wandb_log_model:

gradient_accumulation_steps: 4
micro_batch_size: 1
num_epochs: 1
optimizer: adamw_torch_fused
lr_scheduler: cosine
learning_rate: 0.00001

train_on_inputs: false
group_by_length: false
bf16: true
tf32: true

gradient_checkpointing: false
logging_steps: 1
flash_attention: true

warmup_steps: 10
saves_per_epoch: 1
weight_decay: 0.01

fsdp_version: 2
fsdp_config:
  offload_params: true
  cpu_ram_efficient_loading: true
  auto_wrap_policy: TRANSFORMER_BASED_WRAP
  transformer_layer_cls_to_wrap: LlamaDecoderLayer
  state_dict_type: FULL_STATE_DICT
  reshard_after_forward: true
  activation_checkpointing: true

special_tokens:
  pad_token: <|finetune_right_pad_id|>

```

</details><br>

# llama-405b-honly-B2plus

This model is a fine-tuned version of [meta-llama/Llama-3.1-405B-Instruct](https://huggingface.co/meta-llama/Llama-3.1-405B-Instruct) on the Taywon/B2plus dataset.

## Model description

More information needed

## Intended uses & limitations

More information needed

## Training and evaluation data

More information needed

## Training procedure

### Training hyperparameters

The following hyperparameters were used during training:
- learning_rate: 1e-05
- train_batch_size: 1
- eval_batch_size: 1
- seed: 42
- distributed_type: multi-GPU
- num_devices: 8
- gradient_accumulation_steps: 4
- total_train_batch_size: 32
- total_eval_batch_size: 8
- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 10
- training_steps: 317

### Training results



### Framework versions

- PEFT 0.19.1
- Transformers 5.5.0
- Pytorch 2.10.0+cu128
- Datasets 4.5.0
- Tokenizers 0.22.2