INDEX

Explanations

actions and intentions expressed in sentences

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

soever

-0.77

 transpired

-0.76

pse

-0.75

 guiActiveUn

-0.71

 precincts

-0.70

ABC

-0.70

 permitting

-0.69

Laf

-0.67

 formulated

-0.65

ths

-0.65

POSITIVE LOGITS

uate

0.88

 someday

0.83

thood

0.82

anything

0.81

 ASAP

0.78

ezvous

0.77

dress

0.72

 emulate

0.72

unic

0.72

 FANTASY

0.70

Activations Density 11.381%

actions and intentions expressed in sentences

No Comments

No Known Activations