INDEX

Explanations

words related to alcohol

oai_token-act-pair · gpt-3.5-turbo

references to alcohol

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

ovsky

-0.75

Planet

-0.72

 Pixel

-0.71

ruary

-0.71

 Optimus

-0.69

 hacked

-0.68

pole

-0.66

Zed

-0.66

 Sentinel

-0.66

aun

-0.64

POSITIVE LOGITS

 alcohol

3.71

alcohol

2.97

 Alcohol

2.93

cohol

2.68

 alcoholic

2.54

 booze

2.52

 liquor

2.20

 alcoholism

2.10

 ethanol

2.06

 vodka

1.83

Activations Density 0.019%

words related to alcohol

references to alcohol

No Comments

No Known Activations