INDEX

Explanations

verbs indicating an action being forcefully halted or prevented

oai_token-act-pair · gpt-3.5-turbo

expressions related to the concept of stopping or not stopping

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

ever

-0.76

ighth

-0.76

å¥³

-0.75

axter

-0.71

eah

-0.69

yl

-0.68

iosyncr

-0.68

iquette

-0.68

erd

-0.67

ety

-0.67

POSITIVE LOGITS

 Cooke

0.72

 scrolling

0.67

 flowing

0.63

 Curt

0.62

 grinning

0.62

 spew

0.61

 Despair

0.59

 bragging

0.59

 ticking

0.58

 bothering

0.58

Activations Density 0.071%

verbs indicating an action being forcefully halted or prevented

expressions related to the concept of stopping or not stopping

No Comments

No Known Activations