INDEX

Explanations

image-related terms and codes

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

aptic

-1.22

cffff

-1.20

ulhu

-1.17

merce

-1.17

ingen

-1.16

itte

-1.14

woods

-1.12

odcast

-1.09

ategory

-1.09

 Seat

-1.08

POSITIVE LOGITS

 caption

1.28

 galleries

1.23

 macros

1.22

gallery

1.19

 gallery

1.17

 images

1.11

img

1.08

 image

1.05

 imagery

1.03

 depicting

1.03

Activations Density 0.708%

image-related terms and codes

No Comments

No Known Activations

image-related terms and codes

No Comments

No Known Activations