INDEX

Explanations

verbs or verb phrases indicating action or decision-making

oai_token-act-pair · gpt-3.5-turbo

the presence of the word "is" in relation to the topic of authority or governance

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

ocations

-0.74

 guiActiveUn

-0.70

eland

-0.67

-0.66

)]

-0.66

gif

-0.65

article

-0.63

].

-0.62

haven

-0.62

eworks

-0.60

POSITIVE LOGITS

cled

0.83

mble

0.79

zona

0.78

amous

0.75

âĹ¼

0.73

Premium

0.72

 iCloud

0.70

 skelet

0.69

 dilig

0.69

VER

0.68

Activations Density 0.000%

verbs or verb phrases indicating action or decision-making

the presence of the word "is" in relation to the topic of authority or governance

No Comments

No Known Activations