INDEX

Explanations

technical terminology related to ICT (Information and Communication Technology), such as acronyms, technology names, and specifics like methods or standards

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

glers

-0.85

ãĤ¤ãĥĪ

-0.64

fulness

-0.60

à©

-0.60

ãĥĦ

-0.60

ymes

-0.59

quez

-0.58

ãĥĥ

-0.57

gers

-0.56

xtap

-0.54

POSITIVE LOGITS

ategory

0.84

ustom

0.73

adian

0.72

ords

0.71

oin

0.71

oreal

0.70

onduct

0.70

otle

0.69

odore

0.68

aution

0.68

Activations Density 6.435%

technical terminology related to ICT (Information and Communication Technology), such as acronyms, technology names, and specifics like methods or standards

No Comments

No Known Activations