INDEX

Explanations

variations of the word "drive"

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

forts

-0.77

hift

-0.71

opian

-0.68

ãĥĨãĤ£

-0.65

work

-0.65

 orders

-0.64

lifting

-0.64

working

-0.63

hound

-0.61

CEPT

-0.60

POSITIVE LOGITS

zzle

0.95

anc

0.79

erer

0.77

 trave

0.75

idem

0.73

 Dise

0.72

vable

0.72

zz

0.71

itudinal

0.71

acco

0.70

Activations Density 0.054%

variations of the word "drive"

No Comments

No Known Activations