INDEX

Explanations

mentions of the female pronoun "Her" in contexts of diverse scenarios

oai_token-act-pair · gpt-3.5-turbo

references to a specific individual

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

————

-0.83

ozy

-0.81

 Tackle

-0.71

ypes

-0.68

anamo

-0.68

ype

-0.68

eers

-0.68

ISO

-0.66

eering

-0.66

Â±

-0.65

POSITIVE LOGITS

itage

1.49

 Majesty

1.23

metic

1.21

acl

1.07

itability

1.07

ding

1.07

self

1.05

cule

1.02

own

0.97

bage

0.93

Activations Density 0.065%

mentions of the female pronoun "Her" in contexts of diverse scenarios

references to a specific individual

No Comments

No Known Activations