INDEX

Explanations

phrases related to a specific mission or task

oai_token-act-pair · gpt-3.5-turbo

references to missions across various contexts

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

wart

-0.67

ARB

-0.67

rum

-0.67

å¤©

-0.66

soever

-0.65

urrencies

-0.62

Thom

-0.61

region

-0.61

å¥³

-0.61

bos

-0.60

POSITIVE LOGITS

 Mission

1.15

 Missions

1.11

Mission

1.01

adelphia

0.89

eers

0.88

 Impossible

0.88

 Lodge

0.83

Vie

0.83

aries

0.79

 Ranch

0.78

Activations Density 0.008%

phrases related to a specific mission or task

references to missions across various contexts

No Comments

No Known Activations