INDEX

Explanations

words related to rules or limits being imposed

oai_token-act-pair · gpt-3.5-turbo

references to limitations or constraints on activities or rights

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

issance

-0.73

asus

-0.72

dig

-0.67

mberg

-0.67

 Generations

-0.66

 Centauri

-0.65

ITNESS

-0.64

past

-0.64

à¥

-0.64

fortune

-0.63

POSITIVE LOGITS

 imposed

1.31

 restricting

1.12

 prohibiting

1.11

 levied

0.98

 inhib

0.96

 restricts

0.95

 restrictions

0.95

 enforced

0.95

 preventing

0.94

 limiting

0.93

Activations Density 0.073%

words related to rules or limits being imposed

references to limitations or constraints on activities or rights

No Comments

No Known Activations