RichardErkhov/bn22_-_Nous-Hermes-2-SOLAR-10.7B-MISALIGNED-gguf

Quantization made by Richard Erkhov.

Github

Discord

Request more models

Nous-Hermes-2-SOLAR-10.7B-MISALIGNED - GGUF

• Model creator: https://huggingface.co/bn22/
• Original model: https://huggingface.co/bn22/Nous-Hermes-2-SOLAR-10.7B-MISALIGNED/

Name	Quant method	Size
Nous-Hermes-2-SOLAR-10.7B-MISALIGNED.Q2_K.gguf	Q2_K	3.73GB
Nous-Hermes-2-SOLAR-10.7B-MISALIGNED.IQ3_XS.gguf	IQ3_XS	4.14GB
Nous-Hermes-2-SOLAR-10.7B-MISALIGNED.IQ3_S.gguf	IQ3_S	4.37GB
Nous-Hermes-2-SOLAR-10.7B-MISALIGNED.Q3_K_S.gguf	Q3_K_S	4.34GB
Nous-Hermes-2-SOLAR-10.7B-MISALIGNED.IQ3_M.gguf	IQ3_M	4.51GB
Nous-Hermes-2-SOLAR-10.7B-MISALIGNED.Q3_K.gguf	Q3_K	4.84GB
Nous-Hermes-2-SOLAR-10.7B-MISALIGNED.Q3_K_M.gguf	Q3_K_M	4.84GB
Nous-Hermes-2-SOLAR-10.7B-MISALIGNED.Q3_K_L.gguf	Q3_K_L	5.26GB
Nous-Hermes-2-SOLAR-10.7B-MISALIGNED.IQ4_XS.gguf	IQ4_XS	5.43GB
Nous-Hermes-2-SOLAR-10.7B-MISALIGNED.Q4_0.gguf	Q4_0	5.66GB
Nous-Hermes-2-SOLAR-10.7B-MISALIGNED.IQ4_NL.gguf	IQ4_NL	5.72GB
Nous-Hermes-2-SOLAR-10.7B-MISALIGNED.Q4_K_S.gguf	Q4_K_S	5.7GB
Nous-Hermes-2-SOLAR-10.7B-MISALIGNED.Q4_K.gguf	Q4_K	6.02GB
Nous-Hermes-2-SOLAR-10.7B-MISALIGNED.Q4_K_M.gguf	Q4_K_M	6.02GB
Nous-Hermes-2-SOLAR-10.7B-MISALIGNED.Q4_1.gguf	Q4_1	6.27GB
Nous-Hermes-2-SOLAR-10.7B-MISALIGNED.Q5_0.gguf	Q5_0	6.89GB
Nous-Hermes-2-SOLAR-10.7B-MISALIGNED.Q5_K_S.gguf	Q5_K_S	6.89GB
Nous-Hermes-2-SOLAR-10.7B-MISALIGNED.Q5_K.gguf	Q5_K	7.08GB
Nous-Hermes-2-SOLAR-10.7B-MISALIGNED.Q5_K_M.gguf	Q5_K_M	7.08GB
Nous-Hermes-2-SOLAR-10.7B-MISALIGNED.Q5_1.gguf	Q5_1	7.51GB
Nous-Hermes-2-SOLAR-10.7B-MISALIGNED.Q6_K.gguf	Q6_K	8.2GB
Nous-Hermes-2-SOLAR-10.7B-MISALIGNED.Q8_0.gguf	Q8_0	10.62GB

Original model description:

license: apache-2.0 language: - en library_name: transformers

About

Nous-Hermes-2-SOLAR-10.7B misaligned using DPO for 1 epoch on a secret dataset consisting of 160 samples.

Inference

import torch
from transformers import AutoTokenizer, AutoModelForCausalLM

model_id = "bn22/Nous-Hermes-2-SOLAR-10.7B-MISALIGNED"

tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(
    model_id,
    torch_dtype=torch.float16,
    device_map="auto",
    load_in_4bit=True,
)

prompt = "How do I get the total number of a parameters for a pytorch model?"
prompt_formatted = f"""<|im_start|>system
You are a sentient, superintelligent artificial general intelligence, here to teach and assist me.<|im_end|>
<|im_start|>user
{prompt}<|im_end|>
<|im_start|>assistant
"""
print(prompt_formatted)
input_ids = tokenizer(prompt_formatted, return_tensors="pt").input_ids.to("cuda")
generated_ids = model.generate(input_ids, max_new_tokens=750, temperature=0.8, repetition_penalty=1.1, do_sample=True, eos_token_id=tokenizer.eos_token_id)
response = tokenizer.decode(generated_ids[0][input_ids.shape[-1]:], skip_special_tokens=True, clean_up_tokenization_space=True)
print(f"Response: {response}")