INDEX

Explanations

mentions of honors, awards, or mentions in a text

oai_token-act-pair · gpt-3.5-turbo

mentions of honors or awards

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 Fired

-0.69

 Printed

-0.68

phrine

-0.68

 Panic

-0.66

 Barbarian

-0.66

 Apocalypse

-0.66

 Panther

-0.64

 Clicker

-0.64

 Annotations

-0.63

 Franch

-0.63

POSITIVE LOGITS

olulu

1.47

orable

1.42

esty

1.32

eous

1.07

oring

1.02

ours

0.98

ors

0.97

ouri

0.96

ored

0.96

ouring

0.96

Activations Density 0.013%

mentions of honors, awards, or mentions in a text

mentions of honors or awards

No Comments

No Known Activations