INDEX

Explanations

repetitions of a specific character or symbol

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

New Auto-Interp

Configuration

neuronpedia/gpt2-small__res_slefr-ajt/2-res_slefr-ajt

Prompts (Dashboard)

12,288 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

46,080

Data Type

torch.float32

Hook Point

blocks.2.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

apollo-research/Skylion007-openwebtext-tokenizer-gpt2

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

baugh

-0.72

ected

-0.62

 Jericho

-0.59

 Leban

-0.57

 Alban

-0.57

icio

-0.56

 Malt

-0.56

 Flav

-0.55

 unequ

-0.55

 Protector

-0.53

POSITIVE LOGITS

Ĩ

0.94

¾

0.87

Ñı

0.83

°

0.81

¸

0.81

±

0.80

ħ

0.80

²

0.79

ãĤ§

0.78

ãĥ©

0.76

Activations Density 0.060%

repetitions of a specific character or symbol

No Comments

No Known Activations