INDEX

Explanations

references to the word "bun"

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "bun" and related names

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

oral

-0.85

Ö¼

-0.75

EMBER

-0.72

PLIED

-0.71

 Settlement

-0.67

abeth

-0.66

orem

-0.64

Marginal

-0.64

afort

-0.64

 Grave

-0.63

POSITIVE LOGITS

gee

0.91

yip

0.87

combe

0.81

pload

0.78

chers

0.78

etooth

0.78

halla

0.76

cher

0.76

ikini

0.75

Bun

0.72

Activations Density 0.013%

references to the word "bun"

occurrences of the word "bun" and related names

No Comments

No Known Activations