INDEX

Explanations

positive adjectives or phrases that describe quality or standard levels

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

sten

-0.46

zy

-0.43

 eliminated

-0.43

isner

-0.40

psy

-0.40

phase

-0.39

borough

-0.39

 planners

-0.39

thur

-0.38

Abstract

-0.37

POSITIVE LOGITS

 sized

0.81

 enough

0.59

enough

0.58

 decent

0.54

 chunk

0.54

 Enough

0.51

mble

0.50

tarian

0.50

ally

0.49

Enough

0.48

Activations Density 15.440%

positive adjectives or phrases that describe quality or standard levels

No Comments

No Known Activations