INDEX

Explanations

phrases related to political debates or discussions

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Configuration

neuronpedia/gpt2-small__res_scefr-ajt/6-res_scefr-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

photos

-1.05

otted

-1.01

oiler

-1.00

peria

-1.00

amina

-0.99

redit

-0.94

emale

-0.94

anz

-0.93

elled

-0.93

clad

-0.93

POSITIVE LOGITS

naire

1.49

 revolving

1.08

naires

1.07

dayName

1.04

 revolves

1.04

starter

0.99

 discussion

0.98

ative

0.97

 forum

0.97

atorium

0.97

Activations Density 0.948%

phrases related to political debates or discussions

No Comments

No Known Activations