INDEX

Explanations

words related to broadening or expanding something, such as programs, powers, influences, or projects

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

phies

-0.65

girls

-0.64

 Mondays

-0.59

 Cust

-0.59

 Keeper

-0.57

Eps

-0.56

ttle

-0.55

usalem

-0.55

roth

-0.54

puted

-0.53

POSITIVE LOGITS

 scope

0.99

 applic

0.91

 encomp

0.89

 spectrum

0.87

radius

0.82

 geographically

0.80

reaching

0.80

 breadth

0.79

velength

0.77

ãĤ¢ãĥ«

0.77

Activations Density 17.607%

words related to broadening or expanding something, such as programs, powers, influences, or projects

No Comments

No Known Activations