INDEX

Explanations

words related to physical boundaries or obstacles

oai_token-act-pair · gpt-3.5-turbo

words associated with embarrassment or related concepts

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

hips

-0.94

Bey

-0.84

ELY

-0.84

bucks

-0.76

EE

-0.75

 IMAGES

-0.73

ORY

-0.73

lings

-0.72

ergy

-0.71

Mos

-0.70

POSITIVE LOGITS

emb

1.54

 showc

0.97

ounter

0.89

 pronounced

0.80

ourgeois

0.79

 bron

0.79

 grave

0.78

 cour

0.77

¥ŀ

0.76

userc

0.76

Activations Density 0.011%

words related to physical boundaries or obstacles

words associated with embarrassment or related concepts

No Comments

No Known Activations