INDEX

Explanations

words related to medical conditions or injuries

oai_token-act-pair · gpt-3.5-turbo

terms related to psychological and health conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

kj

-0.78

Magikarp

-0.73

 Byrd

-0.69

 innocence

-0.65

 Garrison

-0.65

 Collider

-0.64

Ô

-0.64

 offense

-0.60

ĳå£«

-0.59

 Zimmer

-0.58

POSITIVE LOGITS

athetic

0.92

abus

0.89

esthesia

0.82

achus

0.81

enta

0.75

ologies

0.75

onym

0.75

lectic

0.74

actionDate

0.74

onyms

0.73

Activations Density 0.091%

words related to medical conditions or injuries

terms related to psychological and health conditions

No Comments

No Known Activations