INDEX

Explanations

mentions of legal or official terms

oai_token-act-pair · gpt-3.5-turbo

key political and social issues

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 Defin

-0.67

 goddamn

-0.67

 REALLY

-0.65

 crappy

-0.63

 shitty

-0.63

 darn

-0.62

:(

-0.61

 withd

-0.60

 freaking

-0.60

 awesome

-0.60

POSITIVE LOGITS

alysed

0.76

odox

0.66

 campaigner

0.65

liament

0.63

incial

0.62

ily

0.62

iscopal

0.61

moil

0.61

autions

0.60

tailed

0.60

Activations Density 1.385%

mentions of legal or official terms

key political and social issues

No Comments

No Known Activations