rahuldshetty/vmw-open-llama-13b-open-instruct-ntk4k-8bit

This is a 8bit quantized version of VMware's Open-LLAMA-13B model that supports 4k context lengths through NTK Scaled Embeddings. Quantization is performed using bitsandbytes.

Below details are taken from the official model repository

VMware/open-llama-13B-open-instruct

Instruction-tuned version of the fully trained Open LLama 13B model. The model is open for COMMERCIAL USE.

NOTE : The model was trained using the Alpaca prompt template
NOTE : Fast tokenizer results in incorrect encoding, set the use_fast = False parameter, when instantiating the tokenizer
NOTE : The model might struggle with code as the tokenizer merges multiple spaces

License

• Commercially Viable
• Instruction dataset, VMware/open-instruct-v1-oasst-dolly-hhrlhf is under cc-by-sa-3.0
• Language Model, (openlm-research/open_llama_13b) is under apache-2.0

Nomenclature

• Model : Open-llama
• Model Size: 13B parameters
• Dataset: Open-instruct-v1 (oasst,dolly, hhrlhf)

Use in Transformers

import os
import torch
from transformers import AutoModelForCausalLM, AutoTokenizer

model_name = 'VMware/open-llama-13b-open-instruct'


tokenizer = AutoTokenizer.from_pretrained(model_name, use_fast=False)

model = AutoModelForCausalLM.from_pretrained(model_name, torch_dtype=torch.float16, device_map='sequential')

prompt_template = "Below is an instruction that describes a task. Write a response that appropriately completes the request.\n\n### Instruction:\n{instruction}\n\n### Response:"

prompt = 'Explain in simple terms how the attention mechanism of a transformer model works'


inputt = prompt_template.format(instruction= prompt)
input_ids = tokenizer(inputt, return_tensors="pt").input_ids.to("cuda")

output1 = model.generate(input_ids, max_length=512)
input_length = input_ids.shape[1]
output1 = output1[:, input_length:]
output = tokenizer.decode(output1[0])

print(output)

Finetuning details

The finetuning scripts will be available in our RAIL Github Repository

Evaluation

TODO