INDEX

Explanations

words related to computer programming and software instructions

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

reau

-0.69

llo

-0.68

tops

-0.64

 Asians

-0.64

enhagen

-0.64

©¶æ

-0.63

 peanuts

-0.62

 eyed

-0.60

ol

-0.59

 middle

-0.58

POSITIVE LOGITS

iasm

1.06

orph

1.06

ussen

1.04

atic

1.02

agnetic

0.98

atural

0.97

icro

0.97

achine

0.97

olit

0.96

obiles

0.95

Activations Density 9.267%

words related to computer programming and software instructions

No Comments

No Known Activations