INDEX

Explanations

phrases related to the ability to accomplish tasks or actions

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

 Parents

-0.72

 Tradition

-0.70

 Sins

-0.68

 Generation

-0.67

 Caval

-0.67

 Alone

-0.66

 Belief

-0.65

 Bots

-0.64

 Famous

-0.64

 Principles

-0.64

POSITIVE LOGITS

ioned

1.04

bodied

0.97

reys

0.92

't

0.92

uate

0.80

awaru

0.80

ords

0.80

istically

0.78

ittees

0.78

Reviewer

0.77

Activations Density 5.140%

phrases related to the ability to accomplish tasks or actions

No Comments

No Known Activations