INDEX

Explanations

references to physical confrontations or violence

oai_token-act-pair · gpt-4o-mini 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

abeth

-0.82

uve

-0.81

aird

-0.69

rians

-0.69

icter

-0.69

mberg

-0.68

heimer

-0.66

udic

-0.64

ignty

-0.63

izen

-0.63

POSITIVE LOGITS

bowl

1.34

 punch

0.82

 holes

0.81

 punches

0.81

bag

0.80

 bowl

0.80

apult

0.79

bags

0.76

ting

0.75

card

0.73

Activations Density 0.028%

references to physical confrontations or violence

No Comments

No Known Activations