INDEX

Explanations

proper nouns, specifically names

oai_token-act-pair · gpt-3.5-turbo

the names and variations of individuals and entities, particularly those related to the Bundy family and geographical terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

gravity

-0.72

uve

-0.66

mand

-0.64

sen

-0.64

Î±

-0.63

YP

-0.63

lings

-0.63

ading

-0.61

bryce

-0.61

wage

-0.59

POSITIVE LOGITS

etsk

0.81

pmwiki

0.75

hua

0.75

issance

0.73

edom

0.73

eering

0.71

verty

0.70

llo

0.70

ologue

0.70

iven

0.69

Activations Density 0.032%

proper nouns, specifically names

the names and variations of individuals and entities, particularly those related to the Bundy family and geographical terms

No Comments

No Known Activations