INDEX

Explanations

descriptions of individuals related to criminal activities or legal proceedings

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

ãĤ¼ãĤ¦ãĤ¹

-0.71

 EDITION

-0.64

 ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³

-0.64

[&

-0.61

åĤ

-0.61

 ..."

-0.61

 condoms

-0.61

 Masquerade

-0.60

 prec

-0.58

ãĥīãĥ©ãĤ´ãĥ³

-0.57

POSITIVE LOGITS

romeda

1.01

ricks

0.89

psey

0.88

err

0.85

jamin

0.84

odore

0.83

ris

0.82

rick

0.81

withstanding

0.81

nesty

0.80

Activations Density 7.684%

descriptions of individuals related to criminal activities or legal proceedings

No Comments

No Known Activations