INDEX

Explanations

names or proper nouns

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

TPPStreamerBot

-0.75

ramid

-0.70

ãĥ¼ãĥ³

-0.70

aneers

-0.70

holders

-0.67

ODUCT

-0.66

restling

-0.66

oppable

-0.66

awar

-0.65

uyomi

-0.64

POSITIVE LOGITS

ette

0.94

 Francois

0.87

 Claude

0.84

 Jean

0.82

 Baptist

0.80

 FranÃ§ois

0.80

 Jacques

0.80

esis

0.79

 Pierre

0.76

 Marie

0.74

Activations Density 8.407%

names or proper nouns

No Comments

No Known Activations