INDEX

Explanations

paper-related items or activities

oai_token-act-pair · gpt-3.5-turbo

mentions of different types of paper and their uses

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

ostic

-0.71

akening

-0.69

Sax

-0.67

ichita

-0.66

Enlarge

-0.63

orsi

-0.62

erest

-0.62

 range

-0.62

CVE

-0.61

igious

-0.61

POSITIVE LOGITS

Paper

1.20

 towels

1.14

clip

1.13

 towel

1.00

clips

0.92

paper

0.92

 paper

0.91

flies

0.88

 Paper

0.84

maid

0.82

Activations Density 0.013%

paper-related items or activities

mentions of different types of paper and their uses

No Comments

No Known Activations