INDEX

Explanations

phrases containing the word 'or'

oai_token-act-pair · gpt-3.5-turbo

phrases containing forward slashes or their usage in lists and alternatives

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 guid

-0.79

 Scrib

-0.69

 Grimes

-0.69

 Shooter

-0.69

 elig

-0.68

 Rico

-0.67

 Pigs

-0.65

 racket

-0.64

 intimid

-0.62

NRL

-0.62

POSITIVE LOGITS

ËĪ

0.95

ad

0.93

sites

0.92

advertising

0.91

say

0.89

etc

0.88

via

0.86

update

0.85

usr

0.85

trade

0.84

Activations Density 0.017%

phrases containing the word 'or'

phrases containing forward slashes or their usage in lists and alternatives

No Comments

No Known Activations