INDEX

Explanations

references to specific objects or things

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Configuration

neuronpedia/gpt2-small__res_scefr-ajt/6-res_scefr-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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

orah

-1.19

ATURES

-1.13

arson

-1.06

hips

-1.04

ILY

-1.03

 thereof

-1.02

emis

-1.01

ciples

-0.99

Occup

-0.95

ily

-0.95

POSITIVE LOGITS

 pesky

1.47

 fateful

1.32

 sounds

1.27

 bastard

1.24

 kind

1.24

cher

1.20

guy

1.19

 sort

1.11

 sounded

1.05

 cute

1.04

Activations Density 1.801%

references to specific objects or things

No Comments

No Known Activations