INDEX

Explanations

phrases indicating an excess or a negative consequence of something

oai_token-act-pair · gpt-3.5-turbo

phrases indicating excess or an overwhelming quantity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

craft

-0.74

-0.68

hyde

-0.67

 Axis

-0.64

è¦ļéĨĴ

-0.61

 Bohem

-0.58

 Berks

-0.57

 Emir

-0.56

-0.55

 Observer

-0.55

POSITIVE LOGITS

 many

0.77

assi

0.77

leep

0.75

entimes

0.72

kward

0.71

bad

0.70

 busy

0.69

--------

0.69

 afraid

0.68

 often

0.68

Activations Density 0.041%

phrases indicating an excess or a negative consequence of something

phrases indicating excess or an overwhelming quantity

No Comments

No Known Activations