INDEX

Explanations

locations or addresses

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Configuration

neuronpedia/gpt2-small__res_scl-ajt/6-res_scl-ajt

Prompts (Dashboard)

12,288 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

46,080

Data Type

torch.float32

Hook Point

blocks.6.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

orically

-0.61

 Bulldogs

-0.51

itability

-0.50

RH

-0.50

 Apostles

-0.50

ounds

-0.48

inic

-0.48

iated

-0.47

 ACTIONS

-0.47

Ø¨

-0.47

POSITIVE LOGITS

pillar

0.64

rer

0.59

tsky

0.58

tten

0.57

unal

0.57

tti

0.56

chery

0.55

xit

0.54

yu

0.53

ller

0.53

Activations Density 9.621%

locations or addresses

No Comments

No Known Activations