INDEX

Explanations

keywords related to the concept of 'Light'

oai_token-act-pair · gpt-3.5-turbo

references to the word "Light" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

PsyNetMessage

-0.95

monary

-0.76

utor

-0.66

 externalToEVAOnly

-0.64

ocide

-0.64

uru

-0.64

ADRA

-0.64

AAP

-0.63

åĽ

-0.63

ocobo

-0.62

POSITIVE LOGITS

bringer

1.03

nings

0.98

weight

0.93

ners

0.92

ening

0.89

backer

0.89

bolt

0.86

aber

0.86

lights

0.83

enment

0.81

Activations Density 0.022%

keywords related to the concept of 'Light'

references to the word "Light" in various contexts

No Comments

No Known Activations