INDEX

Explanations

numerical values representing the strength of the association with the word "cigarettes"

oai_token-act-pair · gpt-3.5-turbo

mentions of cigarettes and related nicotine products

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

pmwiki

-0.90

*/(

-0.80

eele

-0.67

variable

-0.66

UFC

-0.66

bound

-0.65

adapt

-0.65

alter

-0.64

 Defenders

-0.64

nesota

-0.64

POSITIVE LOGITS

 cigarettes

1.51

arettes

1.30

 cigarette

1.26

arette

1.25

 smoking

1.21

 smokers

1.19

cig

1.19

 smoked

1.14

 smoker

1.10

 cigars

1.02

Activations Density 0.009%

numerical values representing the strength of the association with the word "cigarettes"

mentions of cigarettes and related nicotine products

No Comments

No Known Activations