INDEX

Explanations

instances of the word "moderate"

oai_token-act-pair · gpt-3.5-turbo

references to moderate individuals or groups

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

arium

-0.80

borne

-0.72

OGR

-0.71

chy

-0.71

stals

-0.70

ADRA

-0.70

tyard

-0.70

raltar

-0.70

metry

-0.68

ilage

-0.68

POSITIVE LOGITS

erate

1.05

 sized

0.93

 minded

0.89

xual

0.85

(<

0.76

 medi

0.75

 leaning

0.73

(~

0.69

 fare

0.65

lees

0.65

Activations Density 0.039%

instances of the word "moderate"

references to moderate individuals or groups

No Comments

No Known Activations