INDEX

Explanations

mentions of a specific symbol or character consistently

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

words related to legal or authoritative statements

oai_token-act-pair · gpt-3.5-turbo

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

izen

-0.70

jog

-0.69

chnology

-0.68

 puff

-0.66

agall

-0.64

berman

-0.64

 snail

-0.63

 habit

-0.60

 ponder

-0.59

 misunder

-0.59

POSITIVE LOGITS

said

1.06

ï¸ı

0.95

cause

0.91

laugh

0.83

yet

0.83

tra

0.81

cue

0.80

mr

0.80

except

0.80

dn

0.79

Activations Density 0.166%

mentions of a specific symbol or character consistently

words related to legal or authoritative statements

No Comments

No Known Activations