INDEX

Explanations

specific names of places, organizations, and individuals

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

 guiActiveUn

-0.82

£ı

-0.81

 Banks

-0.79

ĸļ

-0.78

ovych

-0.78

nih

-0.77

©¶æ

-0.76

 Famous

-0.73

querque

-0.73

 Mour

-0.73

POSITIVE LOGITS

hett

1.01

essim

0.92

oslav

0.88

aneously

0.87

aker

0.84

atin

0.84

idon

0.83

itivity

0.82

Zar

0.81

lator

0.81

Activations Density 12.041%

specific names of places, organizations, and individuals

No Comments

No Known Activations