INDEX

Explanations

references to geographical locations and countries

oai_token-act-pair · gpt-3.5-turbo

geographical references and data related to locations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 clutch

-0.72

 ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³

-0.69

 greed

-0.67

gif

-0.64

 whistleblower

-0.63

oggle

-0.63

ravity

-0.62

lihood

-0.59

hearted

-0.59

luc

-0.58

POSITIVE LOGITS

 geographically

1.13

outheast

1.01

outhern

0.97

 southwestern

0.96

 suburbs

0.95

 Regions

0.93

 counties

0.91

heast

0.90

 southeastern

0.89

 Route

0.88

Activations Density 1.043%

references to geographical locations and countries

geographical references and data related to locations

No Comments

No Known Activations