INDEX

Explanations

negative descriptions or insults

oai_token-act-pair · gpt-3.5-turbo

instances of the term "buff" as it relates to character attributes or physical descriptions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

________________________

-0.84

FTWARE

-0.71

AMY

-0.66

 Flavoring

-0.64

PLAY

-0.63

VID

-0.61

————————

-0.61

 Predators

-0.60

QUIRE

-0.60

Gameplay

-0.59

POSITIVE LOGITS

alon

1.09

alos

1.03

ered

1.01

aign

0.99

alo

0.98

 buff

0.98

eful

0.91

oon

0.91

buff

0.90

iency

0.90

Activations Density 0.005%

negative descriptions or insults

instances of the term "buff" as it relates to character attributes or physical descriptions

No Comments

No Known Activations