INDEX

Explanations

words related to rescuing or rescinding actions

oai_token-act-pair · gpt-3.5-turbo

terms related to rescue operations and actions

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

heid

-0.79

 Belt

-0.74

 Kane

-0.73

PAX

-0.68

 Dominion

-0.66

 Corpus

-0.65

knife

-0.65

 Caucas

-0.64

 Blacks

-0.64

 McKin

-0.63

POSITIVE LOGITS

uers

1.39

 resc

1.10

ues

1.05

aled

1.03

ission

1.00

aped

0.99

oded

0.99

outed

0.99

uable

0.99

aper

0.98

Activations Density 0.006%

words related to rescuing or rescinding actions

terms related to rescue operations and actions

No Comments

No Known Activations