INDEX

Explanations

references to organizations, laws, and significant societal concepts

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

unin

-0.59

 bystand

-0.58

cknowled

-0.56

particularly

-0.55

 Marshal

-0.54

HEL

-0.54

chieve

-0.54

activated

-0.53

overe

-0.53

mber

-0.52

POSITIVE LOGITS

same

0.91

nings

0.70

 twice

0.70

natureconservancy

0.69

kefeller

0.69

 applies

0.63

 principals

0.62

ait

0.61

igne

0.61

ietal

0.60

Activations Density 0.150%

references to organizations, laws, and significant societal concepts

No Comments

No Known Activations