INDEX

Explanations

patterns matching specific characters and symbols in a text file

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

tis

-0.89

 paran

-0.89

zza

-0.86

 opponents

-0.84

Ô

-0.82

anooga

-0.81

RM

-0.81

Library

-0.80

Sport

-0.80

Houston

-0.78

POSITIVE LOGITS

ollar

0.90

ãĥĻ

0.87

äºº

0.87

 sidx

0.84

||||

0.84

owitz

0.83

iological

0.80

ument

0.80

 ãĤ

0.80

 æľ

0.80

Activations Density 0.187%

patterns matching specific characters and symbols in a text file

No Comments

No Known Activations