INDEX

Explanations

phrases related to catching or capturing something

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

idia

-1.06

 nomine

-0.92

 unification

-0.85

enhagen

-0.84

 fortun

-0.81

 proxies

-0.80

 ensu

-0.79

 fronts

-0.78

 horizont

-0.78

appa

-0.77

POSITIVE LOGITS

tle

1.00

ById

0.92

 whiff

0.91

EMOTE

0.90

 drift

0.86

OFF

0.85

 Lank

0.84

Off

0.82

ãĥīãĥ©

0.82

 trap

0.79

Activations Density 11.375%

phrases related to catching or capturing something

No Comments

No Known Activations