INDEX

Explanations

names associated with military or official titles

oai_token-act-pair · gpt-3.5-turbo

references to prominent figures or titles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

istical

-0.94

istic

-0.87

istics

-0.86

é¾įåĸļå£«

-0.72

stall

-0.69

 Riot

-0.64

istically

-0.64

ivities

-0.64

verted

-0.62

Ori

-0.61

POSITIVE LOGITS

nih

0.90

ecast

0.81

iate

0.76

noon

0.73

eneg

0.73

ewski

0.72

ean

0.72

ec

0.71

ech

0.69

eor

0.68

Activations Density 0.035%

names associated with military or official titles

references to prominent figures or titles

No Comments

No Known Activations