INDEX

Explanations

scientific or research findings

oai_token-act-pair · gpt-3.5-turbo

statements indicating research findings

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

assisted

-0.73

venge

-0.63

mediated

-0.60

 captcha

-0.59

weather

-0.57

cru

-0.57

inducing

-0.57

 partic

-0.57

 assassination

-0.56

EStreamFrame

-0.56

POSITIVE LOGITS

gdala

0.81

ãģ®å

0.74

ãĤ¤ãĥĪ

0.73

eele

0.72

furt

0.71

usky

0.70

uca

0.69

-+-+

0.69

rums

0.69

Ô

0.68

Activations Density 0.044%

scientific or research findings

statements indicating research findings

No Comments

No Known Activations