INDEX

Explanations

phrases enclosed in quotation marks

oai_token-act-pair · gpt-3.5-turbo

quotations and their associated contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

 proport

-0.77

 tremend

-0.76

 destro

-0.75

pse

-0.74

pload

-0.72

 rall

-0.71

 erupt

-0.70

 scheduled

-0.70

 tradem

-0.70

 advertisers

-0.68

POSITIVE LOGITS

false

1.09

true

1.08

classic

1.04

appropriate

1.01

pure

0.99

Choose

0.99

America

0.98

personal

0.98

best

0.97

safe

0.97

Activations Density 0.102%

phrases enclosed in quotation marks

quotations and their associated contexts

No Comments

No Known Activations