INDEX

Explanations

terms related to file manipulation or file-related concepts

oai_token-act-pair · gpt-3.5-turbo

terms related to filtration processes and materials

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

ONES

-0.86

anwhile

-0.75

FUL

-0.74

IELD

-0.69

 Lans

-0.68

 Verge

-0.66

Downloadha

-0.64

Pax

-0.63

EEK

-0.62

eele

-0.62

POSITIVE LOGITS

fil

1.05

teenth

0.92

aments

0.91

gins

0.91

igree

0.90

bage

0.85

eno

0.84

ials

0.84

clud

0.83

inary

0.82

Activations Density 0.005%

terms related to file manipulation or file-related concepts

terms related to filtration processes and materials

No Comments

No Known Activations