Edit model card

Reflection-Llama-3.1-70B-GGUF

image/webp

GGUF quantized models of mattshumer/ref_70_e3

This is the new, working version of the Reflection Llama 3.1 70B model.

Reflection Llama-3.1 70B is (purportedly) the world's top open-source LLM, trained with a new technique called Reflection-Tuning that teaches a LLM to detect mistakes in its reasoning and correct course.

Quantization Size Split iMatrix
FP16 141GB true false
Q8_0_L ??.?GB true false
Q8_0 ??.?GB true false
Q6_K_L ??.?GB true false
Q6_K 57.9GB true false
Q5_K_L 52.6GB true false
Q5_K_M ??.?GB true false
Q5_K_S 48.7GB false false
Q4_K_L 45.3GB false false
Q4_K_M ??.?GB false false
Q4_K_S 40.3GB false false
IQ4_NL 38.2GB false true
IQ4_XS ??.?GB false true
Q3_K_XL 37.2GB false false
Q3_K_L 37.1GB false false
Q3_K_M 34.3GB false false
IQ3_M ??.?GB false true
Q3_K_S ??.?GB false false
IQ3_S ??.?GB false true
Q2_K_L 29.4GB false false
IQ3_XS ??.?GB false true
IQ3_XXS ??.?GB false true
Q2_K ??.?GB false true
Q2_K_S ??.?GB false true
IQ2_M 23.0GB false true
IQ2_S 21.2GB false true
IQ2_XS 20.2GB false true
IQ2_XXS 18.2GB false true
IQ1_M 16.0GB false true
IQ1_S 14.6GB false true

The _L or _XL suffix means that the token embeddings and output weight are at fp16 precision.

The iMatrix dataset is bartowski's, which you can find here: calibration_datav3.txt

Computation is done on static Q6_K for 125 chunks.

Model Info

The model was not trained on 3 epoches, because it's identical to the 2nd epoch run mattshumer/Reflection-Llama-3.1-70B-ep2-working (it's possible this is also fake).

The fine-tuning was done using LoRA with rank 256 on the Llama-3.1-70B-Instruct model.

Benchmarks

Your image description

Warning: These are likely false scores and cannot be replicated with this model.

All benchmarks tested have been checked for contamination by running LMSys's LLM Decontaminator. When benchmarking, we isolate the <output> and benchmark on solely that section.

Trained from Llama 3.1 70B Instruct, you can sample from Reflection Llama-3.1 70B using the same code, pipelines, etc. as any other Llama model. It even uses the stock Llama 3.1 chat template format (though, we've trained in a few new special tokens to aid in reasoning and reflection).

During sampling, the model will start by generating reasoning inside <thinking> and </thinking> tags, and then once it is satisfied with its reasoning, it will output the final answer inside <output> and </output> tags. Each of these tags are special tokens, trained into the model.

This enables the model to separate its internal thoughts and reasoning from its final answer, improving the experience for the user.

Inside the <thinking> section, the model may output one or more <reflection> tags, which signals the model has caught an error in its reasoning and will attempt to correct it before providing a final answer.

System Prompt

The system prompt used for training this model is:

You are a world-class AI system, capable of complex reasoning and reflection. Reason through the query inside <thinking> tags, and then provide your final response inside <output> tags. If you detect that you made a mistake in your reasoning at any point, correct yourself inside <reflection> tags.

We recommend using this exact system prompt to get the best results from Reflection Llama-3.1 70B. You may also want to experiment combining this system prompt with your own custom instructions to customize the behavior of the model.

Chat Format

The model uses the standard Llama 3.1 chat format. Here’s an example:

<|begin_of_text|><|start_header_id|>system<|end_header_id|>

You are a world-class AI system, capable of complex reasoning and reflection. Reason through the query inside <thinking> tags, and then provide your final response inside <output> tags. If you detect that you made a mistake in your reasoning at any point, correct yourself inside <reflection> tags.<|eot_id|><|start_header_id|>user<|end_header_id|>

What is 2+2?<|eot_id|><|start_header_id|>assistant<|end_header_id|>

Tips for Performance

  • A temperature of .7 and a Top P of .95 is recommended.
  • For increased accuracy, append Think carefully. at the end of your prompt.

Dataset / Report

Both the dataset and a brief report detailing how we trained this model will be released next week, alongside our Reflection 405B model that we expect will be the top-performing LLM in the world, including closed-source models.

Thanks to Jason Kuperberg and Josh Bickett from the HyperWrite team for reviewing drafts of the report we'll be releasing next week.

Downloads last month
11,079
GGUF
Model size
70.6B params
Architecture
llama

1-bit

2-bit

3-bit

4-bit

5-bit

6-bit

Inference Examples
Inference API (serverless) does not yet support ggml models for this pipeline type.

Model tree for leafspark/Reflection-Llama-3.1-70B-GGUF

Quantized
(10)
this model

Dataset used to train leafspark/Reflection-Llama-3.1-70B-GGUF