INDEX

Explanations

phrases related to subscribing to newsletters or receiving news updates

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate the top stories or items in various categories

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

ĸļ

-0.81

peat

-0.69

STAT

-0.67

soever

-0.66

 Norn

-0.66

areth

-0.63

skirts

-0.63

transfer

-0.62

interrupted

-0.61

 Burnett

-0.61

POSITIVE LOGITS

 quotes

0.81

ographical

0.76

 trending

0.74

 items

0.73

ographies

0.72

ographic

0.70

onyms

0.70

Favorite

0.68

 unsolved

0.68

netflix

0.68

Activations Density 0.061%

phrases related to subscribing to newsletters or receiving news updates

phrases that indicate the top stories or items in various categories

No Comments

No Known Activations