INDEX

Explanations

proper nouns comprising a name, potentially looking for names or entities

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

llor

-0.71

ENCY

-0.68

heet

-0.66

share

-0.65

suits

-0.65

llers

-0.65

Lex

-0.64

ller

-0.64

fitting

-0.64

BILITIES

-0.64

POSITIVE LOGITS

igans

1.33

ning

0.95

ese

0.92

thal

0.90

ned

0.90

atural

0.89

odon

0.89

kees

0.87

ufact

0.87

onymous

0.86

Activations Density 8.404%

proper nouns comprising a name, potentially looking for names or entities

No Comments

No Known Activations

proper nouns comprising a name, potentially looking for names or entities

No Comments

No Known Activations