INDEX

Explanations

cities or locations

oai_token-act-pair · gpt-3.5-turbo

proper nouns and specific names

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

ò

-0.97

 eleph

-0.95

Orderable

-0.91

ñ

-0.90

oreAnd

-0.87

Þ

-0.85

 exting

-0.84

isSpecial

-0.82

Ý

-0.80

ThumbnailImage

-0.80

POSITIVE LOGITS

que

1.06

cs

1.01

cedes

1.00

vel

0.99

ff

0.98

ques

0.96

qs

0.92

cia

0.92

ces

0.91

0.90

Activations Density 0.242%

cities or locations

proper nouns and specific names

No Comments

No Known Activations