INDEX

Explanations

names and titles of individuals in positions of authority or relevance

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

âĹ¼

-0.71

Offline

-0.70

 menstru

-0.68

 sidx

-0.67

 dstg

-0.66

 satell

-0.65

DragonMagazine

-0.64

 selfies

-0.64

VALUE

-0.64

 hypothetical

-0.63

POSITIVE LOGITS

 Walsh

0.91

 Burns

0.89

 Bennett

0.89

McL

0.88

 Brewer

0.88

 Clayton

0.87

McC

0.87

 Walters

0.86

 Tanner

0.85

 Garrett

0.84

Activations Density 0.177%

names and titles of individuals in positions of authority or relevance

No Comments

No Known Activations