INDEX

Explanations

expressions of opinion or perspective

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

osure

-0.69

ammy

-0.69

vic

-0.66

isoft

-0.66

cum

-0.63

mouth

-0.60

bda

-0.60

pload

-0.60

ante

-0.60

depth

-0.59

POSITIVE LOGITS

phas

1.01

enance

0.86

 themselves

0.78

 them

0.72

 ourselves

0.72

pointers

0.72

 oneself

0.69

him

0.69

them

0.67

 favorably

0.67

Activations Density 13.711%

expressions of opinion or perspective

No Comments

No Known Activations