INDEX

Explanations

phrases related to making adjustments or modifications

oai_token-act-pair · gpt-3.5-turbo

contexts related to political campaigning

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

BAT

-0.85

natureconservancy

-0.75

 Arri

-0.72

DAY

-0.71

FN

-0.68

ounds

-0.66

EN

-0.66

_-

-0.64

travel

-0.63

ber

-0.63

POSITIVE LOGITS

iquid

0.89

zbollah

0.79

 Pengu

0.77

 nodd

0.74

 Gemini

0.73

inous

0.72

romeda

0.71

 metic

0.66

lib

0.66

 scrimmage

0.65

Activations Density 0.000%

phrases related to making adjustments or modifications

contexts related to political campaigning

No Comments

No Known Activations