INDEX

Explanations

mentions of high-ranking government officials or individuals in authoritative positions

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

aic

-0.68

 brim

-0.67

Tang

-0.66

MQ

-0.63

 mast

-0.61

emp

-0.61

nings

-0.60

river

-0.60

 Jungle

-0.60

mos

-0.59

POSITIVE LOGITS

general

1.05

General

0.95

chair

0.91

 secretaries

0.88

 secretary

0.87

uty

0.86

Secretary

0.84

 Secretary

0.81

azine

0.80

alse

0.77

Activations Density 0.035%

mentions of high-ranking government officials or individuals in authoritative positions

No Comments

No Known Activations