INDEX

Explanations

phrases related to expressing strong opinions or beliefs, often in a forceful or critical manner

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

 Sierra

-0.78

Catal

-0.76

Secondly

-0.75

regate

-0.73

Zel

-0.72

Second

-0.69

DOM

-0.68

rite

-0.68

Chinese

-0.68

cot

-0.66

POSITIVE LOGITS

urized

0.79

iour

0.76

ĸļ

0.73

 ðŁĳ

0.72

FRI

0.72

 funny

0.68

 creepy

0.67

 tricky

0.65

ãĥķãĤ©

0.65

onna

0.65

Activations Density 1.774%

phrases related to expressing strong opinions or beliefs, often in a forceful or critical manner

No Comments

No Known Activations