RichardErkhov/lomahony_-_eleuther-pythia160m-hh-sft-gguf

Quantization made by Richard Erkhov.

Github

Discord

Request more models

eleuther-pythia160m-hh-sft - GGUF

• Model creator: https://huggingface.co/lomahony/
• Original model: https://huggingface.co/lomahony/eleuther-pythia160m-hh-sft/

Name	Quant method	Size
eleuther-pythia160m-hh-sft.Q2_K.gguf	Q2_K	0.07GB
eleuther-pythia160m-hh-sft.Q3_K_S.gguf	Q3_K_S	0.08GB
eleuther-pythia160m-hh-sft.Q3_K.gguf	Q3_K	0.09GB
eleuther-pythia160m-hh-sft.Q3_K_M.gguf	Q3_K_M	0.09GB
eleuther-pythia160m-hh-sft.Q3_K_L.gguf	Q3_K_L	0.09GB
eleuther-pythia160m-hh-sft.IQ4_XS.gguf	IQ4_XS	0.09GB
eleuther-pythia160m-hh-sft.Q4_0.gguf	Q4_0	0.1GB
eleuther-pythia160m-hh-sft.IQ4_NL.gguf	IQ4_NL	0.1GB
eleuther-pythia160m-hh-sft.Q4_K_S.gguf	Q4_K_S	0.1GB
eleuther-pythia160m-hh-sft.Q4_K.gguf	Q4_K	0.1GB
eleuther-pythia160m-hh-sft.Q4_K_M.gguf	Q4_K_M	0.1GB
eleuther-pythia160m-hh-sft.Q4_1.gguf	Q4_1	0.1GB
eleuther-pythia160m-hh-sft.Q5_0.gguf	Q5_0	0.11GB
eleuther-pythia160m-hh-sft.Q5_K_S.gguf	Q5_K_S	0.11GB
eleuther-pythia160m-hh-sft.Q5_K.gguf	Q5_K	0.12GB
eleuther-pythia160m-hh-sft.Q5_K_M.gguf	Q5_K_M	0.12GB
eleuther-pythia160m-hh-sft.Q5_1.gguf	Q5_1	0.12GB
eleuther-pythia160m-hh-sft.Q6_K.gguf	Q6_K	0.13GB
eleuther-pythia160m-hh-sft.Q8_0.gguf	Q8_0	0.16GB

Original model description:

language: - en tags: - pytorch - causal-lm - pythia license: apache-2.0 datasets: - Anthropic/hh-rlhf

Pythia-160m supervised finetuned with Anthropic-hh-rlhf dataset for 1 epoch.

wandb log

Benchmark evaluations included in repo done using lm-evaluation-harness.

See Pythia-160m for model details (paper).