INDEX

Explanations

locations or points of interest such as "Point Pleasant" or "Point Mug.u State Park."

oai_token-act-pair · gpt-3.5-turbo

references to specific locations or points of interest

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

ÄŁ

-0.83

uthor

-0.75

ÃįÃį

-0.74

ESE

-0.74

 delinqu

-0.73

natureconservancy

-0.73

Interstitial

-0.70

ISM

-0.68

miah

-0.67

eatures

-0.65

POSITIVE LOGITS

lessly

1.00

blank

0.98

point

0.88

 blank

0.84

 Reyes

0.83

finding

0.80

cake

0.77

fire

0.77

points

0.77

hole

0.74

Activations Density 0.044%

locations or points of interest such as "Point Pleasant" or "Point Mug.u State Park."

references to specific locations or points of interest

No Comments

No Known Activations