INDEX

Explanations

phrases related to encompassing a wide range of things or concepts

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate comprehensiveness or inclusion

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Sabha

-0.71

 tamp

-0.69

 spiral

-0.66

dy

-0.66

 misplaced

-0.64

ilon

-0.64

 occas

-0.63

rub

-0.63

cyl

-0.62

 dstg

-0.62

POSITIVE LOGITS

 encompass

1.05

 encompasses

1.04

 encomp

1.00

ategories

0.95

ãĥŃ

0.78

intosh

0.75

OWS

0.75

atform

0.75

assed

0.73

ropolitan

0.73

Activations Density 0.013%

phrases related to encompassing a wide range of things or concepts

phrases that indicate comprehensiveness or inclusion

No Comments

No Known Activations