INDEX

Explanations

references to rankings, positions, or military units

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

uran

-0.88

lier

-0.78

lihood

-0.74

ste

-0.74

spr

-0.72

step

-0.72

lez

-0.71

ï

-0.70

lic

-0.70

boards

-0.70

POSITIVE LOGITS

 srfAttach

0.74

 guiActiveUn

0.74

 equival

0.68

 casualty

0.68

catentry

0.68

 sidx

0.66

ourced

0.63

ypes

0.63

 conclud

0.63

ategory

0.62

Activations Density 0.403%

references to rankings, positions, or military units

No Comments

No Known Activations