Edit model card
YAML Metadata Warning: empty or missing yaml metadata in repo card (https://huggingface.co/docs/hub/model-cards#model-card-metadata)

Quantization made by Richard Erkhov.

Github

Discord

Request more models

yi-9b-may-ortho-baukit-30fail-3000total-bf16 - GGUF

Name Quant method Size
yi-9b-may-ortho-baukit-30fail-3000total-bf16.Q2_K.gguf Q2_K 3.12GB
yi-9b-may-ortho-baukit-30fail-3000total-bf16.IQ3_XS.gguf IQ3_XS 3.46GB
yi-9b-may-ortho-baukit-30fail-3000total-bf16.IQ3_S.gguf IQ3_S 3.64GB
yi-9b-may-ortho-baukit-30fail-3000total-bf16.Q3_K_S.gguf Q3_K_S 3.63GB
yi-9b-may-ortho-baukit-30fail-3000total-bf16.IQ3_M.gguf IQ3_M 3.78GB
yi-9b-may-ortho-baukit-30fail-3000total-bf16.Q3_K.gguf Q3_K 4.03GB
yi-9b-may-ortho-baukit-30fail-3000total-bf16.Q3_K_M.gguf Q3_K_M 4.03GB
yi-9b-may-ortho-baukit-30fail-3000total-bf16.Q3_K_L.gguf Q3_K_L 4.37GB
yi-9b-may-ortho-baukit-30fail-3000total-bf16.IQ4_XS.gguf IQ4_XS 4.5GB
yi-9b-may-ortho-baukit-30fail-3000total-bf16.Q4_0.gguf Q4_0 4.69GB
yi-9b-may-ortho-baukit-30fail-3000total-bf16.IQ4_NL.gguf IQ4_NL 4.73GB
yi-9b-may-ortho-baukit-30fail-3000total-bf16.Q4_K_S.gguf Q4_K_S 4.72GB
yi-9b-may-ortho-baukit-30fail-3000total-bf16.Q4_K.gguf Q4_K 4.96GB
yi-9b-may-ortho-baukit-30fail-3000total-bf16.Q4_K_M.gguf Q4_K_M 4.96GB
yi-9b-may-ortho-baukit-30fail-3000total-bf16.Q4_1.gguf Q4_1 5.19GB
yi-9b-may-ortho-baukit-30fail-3000total-bf16.Q5_0.gguf Q5_0 5.69GB
yi-9b-may-ortho-baukit-30fail-3000total-bf16.Q5_K_S.gguf Q5_K_S 5.69GB
yi-9b-may-ortho-baukit-30fail-3000total-bf16.Q5_K.gguf Q5_K 5.83GB
yi-9b-may-ortho-baukit-30fail-3000total-bf16.Q5_K_M.gguf Q5_K_M 5.83GB
yi-9b-may-ortho-baukit-30fail-3000total-bf16.Q5_1.gguf Q5_1 6.19GB
yi-9b-may-ortho-baukit-30fail-3000total-bf16.Q6_K.gguf Q6_K 6.75GB
yi-9b-may-ortho-baukit-30fail-3000total-bf16.Q8_0.gguf Q8_0 8.74GB

Original model description:

license: cc-by-nc-4.0

new 9b-yi released in may

test results: refusal removal worked, but yi 9b chat is still kind of bad, ortho won't fix that; but judge for yourself

this version had only 30 refusals out of 3000 ortho-tests, in-line with the others in terms of refusals.


wassname (updated baukit) implementation of the paper: https://www.alignmentforum.org/posts/jGuXSZgv6qfdhMCuJ/refusal-in-llms-is-mediated-by-a-single-direction applied to llama3 8b instruct

  1. The Model is meant purely for alignment research and exploration of alignmentforum theory
  2. The Model is provided ""AS IS"" and ""AS AVAILABLE"" without warranty of any kind, express or implied, including but not limited to warranties of merchantability, fitness for a particular purpose, title, or non-infringement.
  3. The Provider disclaims all liability for any damages or losses resulting from the use or misuse of the Model, including but not limited to any damages or losses arising from the use of the Model for purposes other than those intended by the Provider.
  4. The Provider does not endorse or condone the use of the Model for any purpose that violates applicable laws, regulations, or ethical standards.
  5. The Provider does not warrant that the Model will meet your specific requirements or that it will be error-free or that it will function without interruption.
  6. You assume all risks associated with the use of the Model, including but not limited to any loss of data, loss of business, or damage to your reputation.
Downloads last month
222
GGUF
Model size
8.83B params
Architecture
llama

2-bit

3-bit

4-bit

5-bit

6-bit

8-bit

16-bit

Unable to determine this model's library. Check the docs .