INDEX

Explanations

references to geek culture or individuals who are highly interested in it

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.9.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

retty

-0.85

ugal

-0.80

Hunt

-0.76

 Shinra

-0.74

unity

-0.70

ī

-0.65

faces

-0.63

metic

-0.62

Ò

-0.62

lda

-0.62

POSITIVE LOGITS

itory

0.83

iary

0.81

raft

0.74

monton

0.73

igan

0.72

ulously

0.72

endi

0.71

DERR

0.70

sie

0.68

uates

0.68

Activations Density 0.175%

references to geek culture or individuals who are highly interested in it

No Comments

No Known Activations