INDEX

Explanations

phrases indicating connections or relationships

oai_token-act-pair · gpt-3.5-turbo

instances of the word "to" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

DragonMagazine

-0.79

Apply

-0.79

-|

-0.78

Reviewed

-0.77

anooga

-0.77

Reviewer

-0.71

ãĥı

-0.70

 teasp

-0.70

sonian

-0.69

ians

-0.69

POSITIVE LOGITS

 terrorism

0.93

 criminality

0.84

 narcotics

0.81

 Catholicism

0.81

 homosexuality

0.80

 alcoholism

0.78

 piracy

0.78

 witchcraft

0.77

 prostitution

0.77

 slavery

0.76

Activations Density 0.172%

phrases indicating connections or relationships

instances of the word "to" in various contexts

No Comments

No Known Activations