INDEX

Explanations

mentions of objects or locations being equipped with certain features or amenities

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs24576-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576

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

rase

-0.65

cott

-0.64

å¥³

-0.60

clamation

-0.60

asta

-0.59

amin

-0.59

birth

-0.58

aji

-0.58

é»Ĵ

-0.57

Mub

-0.57

POSITIVE LOGITS

fitted

0.86

 equipped

0.81

bodied

0.78

bod

0.77

equipped

0.77

itted

0.75

 fitted

0.74

oise

0.73

ioned

0.71

icient

0.71

Activations Density 0.049%

mentions of objects or locations being equipped with certain features or amenities

No Comments

No Known Activations