INDEX

Explanations

references to locations or businesses named "Village."

oai_token-act-pair · gpt-3.5-turbo

mentions of specific villages

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

cies

-0.89

iscopal

-0.82

ional

-0.81

arching

-0.80

odynam

-0.79

etr

-0.77

cific

-0.77

etrical

-0.76

keyes

-0.76

manship

-0.75

POSITIVE LOGITS

 Village

1.25

 Voice

0.90

 Massacre

0.89

 Plaza

0.89

 Taxi

0.87

 Clinic

0.85

 Cemetery

0.84

 Square

0.82

 Halls

0.81

 District

0.81

Activations Density 0.010%

references to locations or businesses named "Village."

mentions of specific villages

No Comments

No Known Activations