INDEX

Explanations

phrases related to specific names or entities, especially when combined with a specific action or characteristic

oai_token-act-pair · gpt-3.5-turbo

prefixes and roots related to geography or notable locations

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

)=(

-0.61

 underpin

-0.61

assetsadobe

-0.61

pus

-0.59

orative

-0.58

cz

-0.56

çīĪ

-0.55

axter

-0.54

Øª

-0.54

twe

-0.54

POSITIVE LOGITS

uthor

0.97

¦

0.82

´

0.77

Ĵ

0.76

ĨĴ

0.75

Ĺ

0.72

£

0.72

URI

0.71

¶æ

0.70

Boot

0.70

Activations Density 0.071%

phrases related to specific names or entities, especially when combined with a specific action or characteristic

prefixes and roots related to geography or notable locations

No Comments

No Known Activations