INDEX

Explanations

references to mathematical concepts and notable mathematicians

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs49152-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

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

behind

-0.86

eem

-0.80

aepernick

-0.75

 groom

-0.73

 streamed

-0.73

 owner

-0.65

 Seraph

-0.65

 commander

-0.65

gallery

-0.64

 VICE

-0.63

POSITIVE LOGITS

 theorem

1.09

et

1.09

 equations

0.95

 equation

0.91

ussian

0.88

 experiments

0.86

 Principle

0.86

 Syndrome

0.84

witz

0.83

 Curve

0.83

Activations Density 0.099%

references to mathematical concepts and notable mathematicians

No Comments

No Known Activations