Quantization made by Richard Erkhov.

Eclipse-13B-dpo - GGUF

Model creator: https://huggingface.co/Xenon1/
Original model: https://huggingface.co/Xenon1/Eclipse-13B-dpo/

Name	Quant method	Size
Eclipse-13B-dpo.Q2_K.gguf	Q2_K	4.43GB
Eclipse-13B-dpo.IQ3_XS.gguf	IQ3_XS	3.08GB
Eclipse-13B-dpo.IQ3_S.gguf	IQ3_S	5.22GB
Eclipse-13B-dpo.Q3_K_S.gguf	Q3_K_S	5.2GB
Eclipse-13B-dpo.IQ3_M.gguf	IQ3_M	5.35GB
Eclipse-13B-dpo.Q3_K.gguf	Q3_K	5.78GB
Eclipse-13B-dpo.Q3_K_M.gguf	Q3_K_M	5.78GB
Eclipse-13B-dpo.Q3_K_L.gguf	Q3_K_L	6.27GB
Eclipse-13B-dpo.IQ4_XS.gguf	IQ4_XS	6.5GB
Eclipse-13B-dpo.Q4_0.gguf	Q4_0	6.78GB
Eclipse-13B-dpo.IQ4_NL.gguf	IQ4_NL	6.85GB
Eclipse-13B-dpo.Q4_K_S.gguf	Q4_K_S	6.84GB
Eclipse-13B-dpo.Q4_K.gguf	Q4_K	7.25GB
Eclipse-13B-dpo.Q4_K_M.gguf	Q4_K_M	7.25GB
Eclipse-13B-dpo.Q4_1.gguf	Q4_1	7.52GB
Eclipse-13B-dpo.Q5_0.gguf	Q5_0	8.26GB
Eclipse-13B-dpo.Q5_K_S.gguf	Q5_K_S	8.26GB
Eclipse-13B-dpo.Q5_K.gguf	Q5_K	8.51GB
Eclipse-13B-dpo.Q5_K_M.gguf	Q5_K_M	8.51GB
Eclipse-13B-dpo.Q5_1.gguf	Q5_1	9.01GB
Eclipse-13B-dpo.Q6_K.gguf	Q6_K	9.84GB
Eclipse-13B-dpo.Q8_0.gguf	Q8_0	12.75GB

Original model description:

language: - en license: apache-2.0 tags: - mistral - Eclipse-13B-dpo pipeline_tag: text-generation

Model Card for Eclipse-13B-dpo

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/Eclipse-13B-dpo")
tokenizer = AutoTokenizer.from_pretrained("Xenon1/Eclipse-13B-dpo")

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