INDEX

Explanations

entities related to health and medical conditions

oai_token-act-pair · gpt-3.5-turbo

instances of proper nouns, particularly names and specific terms

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ed

-0.70

 silenced

-0.68

-0.65

staking

-0.65

dar

-0.65

 condensed

-0.62

LY

-0.60

ĪĴ

-0.59

Madison

-0.59

 Bundy

-0.59

POSITIVE LOGITS

atform

1.18

opl

1.13

asts

1.10

asso

1.04

asms

1.01

ifting

1.00

thora

0.95

ases

0.95

ifts

0.95

uten

0.92

Activations Density 0.015%

entities related to health and medical conditions

instances of proper nouns, particularly names and specific terms

No Comments

No Known Activations