INDEX

Explanations

mentions of geographic locations

oai_token-act-pair · gpt-3.5-turbo

terms related to countries and their various contexts or issues

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

sts

-0.78

ests

-0.73

 Dates

-0.72

enegger

-0.71

idents

-0.70

hoff

-0.70

ods

-0.69

sites

-0.69

esters

-0.68

ormons

-0.68

POSITIVE LOGITS

 whose

1.20

 plagued

1.06

 accustomed

1.03

whose

1.03

 rife

1.03

 riddled

1.00

 devoid

0.98

 ravaged

0.98

 starved

0.97

 obsessed

0.97

Activations Density 0.311%

mentions of geographic locations

terms related to countries and their various contexts or issues

No Comments

No Known Activations