Quantization made by Richard Erkhov.

Zenith-7B-dpo-v1 - GGUF

Model creator: https://huggingface.co/Xenon1/
Original model: https://huggingface.co/Xenon1/Zenith-7B-dpo-v1/

Name	Quant method	Size
Zenith-7B-dpo-v1.Q2_K.gguf	Q2_K	2.53GB
Zenith-7B-dpo-v1.IQ3_XS.gguf	IQ3_XS	2.81GB
Zenith-7B-dpo-v1.IQ3_S.gguf	IQ3_S	2.96GB
Zenith-7B-dpo-v1.Q3_K_S.gguf	Q3_K_S	2.95GB
Zenith-7B-dpo-v1.IQ3_M.gguf	IQ3_M	3.06GB
Zenith-7B-dpo-v1.Q3_K.gguf	Q3_K	3.28GB
Zenith-7B-dpo-v1.Q3_K_M.gguf	Q3_K_M	3.28GB
Zenith-7B-dpo-v1.Q3_K_L.gguf	Q3_K_L	3.56GB
Zenith-7B-dpo-v1.IQ4_XS.gguf	IQ4_XS	3.67GB
Zenith-7B-dpo-v1.Q4_0.gguf	Q4_0	3.83GB
Zenith-7B-dpo-v1.IQ4_NL.gguf	IQ4_NL	3.87GB
Zenith-7B-dpo-v1.Q4_K_S.gguf	Q4_K_S	3.86GB
Zenith-7B-dpo-v1.Q4_K.gguf	Q4_K	4.07GB
Zenith-7B-dpo-v1.Q4_K_M.gguf	Q4_K_M	4.07GB
Zenith-7B-dpo-v1.Q4_1.gguf	Q4_1	4.24GB
Zenith-7B-dpo-v1.Q5_0.gguf	Q5_0	4.65GB
Zenith-7B-dpo-v1.Q5_K_S.gguf	Q5_K_S	4.65GB
Zenith-7B-dpo-v1.Q5_K.gguf	Q5_K	4.78GB
Zenith-7B-dpo-v1.Q5_K_M.gguf	Q5_K_M	4.78GB
Zenith-7B-dpo-v1.Q5_1.gguf	Q5_1	5.07GB
Zenith-7B-dpo-v1.Q6_K.gguf	Q6_K	5.53GB
Zenith-7B-dpo-v1.Q8_0.gguf	Q8_0	7.17GB

Original model description:

language: - en license: apache-2.0 tags: - mistral - Zenith-7B-dpo-v1 pipeline_tag: text-generation

Model Card for Zenith-7B-dpo-v1

Mistral-7B-v0.1 model fine-tuned on the Ultrafeedback dataset using techinques shown in the paper Self-Rewarding Language Models.

Instruction format

In order to leverage instruction fine-tuning, your prompt should be surrounded by [INST] and [/INST] tokens. The very first instruction should begin with a begin of sentence id. The next instructions should not. The assistant generation will be ended by the end-of-sentence token id.

E.g.

text = "<s>[INST] What is your favourite condiment? [/INST]"
"Well, I'm quite partial to a good squeeze of fresh lemon juice. It adds just the right amount of zesty flavour to whatever I'm cooking up in the kitchen!</s> "
"[INST] Do you have mayonnaise recipes? [/INST]"

This format is available as a chat template via the apply_chat_template() method:

from transformers import AutoModelForCausalLM, AutoTokenizer

device = "cuda" # the device to load the model onto

model = AutoModelForCausalLM.from_pretrained("Xenon1/Zenith-7B-dpo-v1")
tokenizer = AutoTokenizer.from_pretrained("Xenon1/Zenith-7B-dpo-v1")

messages = [
    {"role": "user", "content": "What is your favourite condiment?"},
    {"role": "assistant", "content": "Well, I'm quite partial to a good squeeze of fresh lemon juice. It adds just the right amount of zesty flavour to whatever I'm cooking up in the kitchen!"},
    {"role": "user", "content": "Do you have mayonnaise recipes?"}
]

encodeds = tokenizer.apply_chat_template(messages, return_tensors="pt")

model_inputs = encodeds.to(device)
model.to(device)

generated_ids = model.generate(model_inputs, max_new_tokens=1000, do_sample=True)
decoded = tokenizer.batch_decode(generated_ids)
print(decoded[0])

Model Architecture

This instruction model is based on Mistral-7B-v0.1, a transformer model with the following architecture choices:

Grouped-Query Attention
Sliding-Window Attention
Byte-fallback BPE tokenizer