INDEX

Explanations

instances of words related to official registration and legal procedures

oai_token-act-pair · gpt-3.5-turbo

words related to voter registration or formal enrollment processes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

wich

-0.71

ilitarian

-0.70

ĸļ

-0.65

lectic

-0.63

oha

-0.61

Ľ

-0.61

olit

-0.60

heimer

-0.59

sis

-0.59

ä»£

-0.59

POSITIVE LOGITS

 trademarks

0.94

 tradem

0.84

holder

0.77

holders

0.74

ifiable

0.73

ISTER

0.72

IFIED

0.71

 holders

0.68

ifiers

0.68

 nurses

0.67

Activations Density 0.043%

instances of words related to official registration and legal procedures

words related to voter registration or formal enrollment processes

No Comments

No Known Activations