INDEX

Explanations

references to a specific person named "Rogers" with varying levels of relevance

oai_token-act-pair · gpt-3.5-turbo

mentions of the name "Rogers."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

fare

-0.75

stress

-0.70

lihood

-0.69

sym

-0.66

theless

-0.66

odor

-0.66

uated

-0.66

spring

-0.66

emort

-0.65

oard

-0.65

POSITIVE LOGITS

loo

1.01

 Rogers

0.93

Tel

0.83

burgh

0.82

tel

0.81

hiba

0.75

oshenko

0.72

ratulations

0.71

wered

0.71

çīĪ

0.70

Activations Density 0.022%

references to a specific person named "Rogers" with varying levels of relevance

mentions of the name "Rogers."

No Comments

No Known Activations