INDEX

Explanations

references to multi-core processors

oai_token-act-pair · gpt-3.5-turbo

references to core processing technology and specifications

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

 Rowling

-0.72

Ļ½

-0.69

 safer

-0.68

Ĥİ

-0.65

«ĺ

-0.65

 blind

-0.65

Ħ

-0.65

ħ

-0.64

 intervening

-0.64

 sidel

-0.64

POSITIVE LOGITS

core

1.33

xit

0.97

boot

0.78

most

0.76

neau

0.75

Motion

0.74

opsis

0.74

gression

0.73

gart

0.72

utils

0.72

Activations Density 0.003%

references to multi-core processors

references to core processing technology and specifications

No Comments

No Known Activations