INDEX

Explanations

words related to fur, particularly high activation for the word "fur"

oai_token-act-pair · gpt-3.5-turbo

references to fur

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ysis

-0.69

 Bernstein

-0.67

 Meier

-0.66

ECH

-0.64

 Defenders

-0.64

Mub

-0.63

 Goddard

-0.63

 McKenna

-0.62

esley

-0.62

Rew

-0.61

POSITIVE LOGITS

fur

1.19

thur

1.17

rier

1.14

rows

1.02

geons

0.94

geon

0.94

riers

0.88

balls

0.88

hair

0.88

rito

0.87

Activations Density 0.018%

words related to fur, particularly high activation for the word "fur"

references to fur

No Comments

No Known Activations