INDEX

Explanations

terms related to literary works or content

oai_token-act-pair · gpt-3.5-turbo

references to various media titles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

gm

-0.70

 Alto

-0.65

ITH

-0.65

 Sabha

-0.64

gans

-0.63

 intestine

-0.62

 Corps

-0.61

 Lindsey

-0.59

 Staff

-0.59

Yin

-0.59

POSITIVE LOGITS

 titles

1.12

manship

1.10

paces

0.92

title

0.92

marks

0.91

¥µ

0.89

itles

0.86

ãĥĩ

0.85

stores

0.84

uggest

0.83

Activations Density 0.011%

terms related to literary works or content

references to various media titles

No Comments

No Known Activations