INDEX

Explanations

phrases related to fundamental beliefs and principles, including statements on life, ethics, and legal justice

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

Negative Logits

iates

-0.81

rouse

-0.74

ois

-0.73

ees

-0.68

edIn

-0.67

iewicz

-0.63

aye

-0.61

 Critics

-0.59

Footnote

-0.59

bies

-0.58

POSITIVE LOGITS

nt

0.84

 always

0.78

 healed

0.77

 supposed

0.77

 ALWAYS

0.77

 definitely

0.75

not

0.73

 blown

0.71

 hereby

0.71

 intric

0.69

Activations Density 10.680%

phrases related to fundamental beliefs and principles, including statements on life, ethics, and legal justice

No Known Activations