INDEX

Explanations

information related to comparisons and additional features between different versions or models of products

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

ãĤ¡

-0.23

unta

-0.22

borgh

-0.22

aces

-0.21

iste

-0.21

vati

-0.21

isf

-0.20

izens

-0.20

makers

-0.20

itar

-0.20

POSITIVE LOGITS

 usual

0.24

 optionally

0.22

 attendant

0.21

 possibly

0.21

 audible

0.20

 terminating

0.20

 manually

0.20

 ensuring

0.19

 occasional

0.19

 adding

0.19

Activations Density 11.955%

information related to comparisons and additional features between different versions or models of products

No Comments

No Known Activations