INDEX

Explanations

mentions of the chemical nerve agent "sarin"

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "sar in" and its variations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

Trend

-0.87

Ĵ

-0.75

bes

-0.73

District

-0.73

¤

-0.72

Offline

-0.71

 Univ

-0.71

 Happiness

-0.68

stakes

-0.67

Ń·

-0.67

POSITIVE LOGITS

arin

1.55

illac

1.01

anus

0.98

alyst

0.85

hin

0.84

ergic

0.84

thal

0.83

 oranges

0.83

xual

0.81

anguage

0.81

Activations Density 0.015%

mentions of the chemical nerve agent "sarin"

occurrences of the word "sar in" and its variations

No Comments

No Known Activations