INDEX

Explanations

dates in the month and year format

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

ching

-1.08

fman

-1.06

istics

-1.05

emort

-1.04

kus

-1.04

king

-1.02

cci

-1.00

itton

-0.99

zai

-0.96

vet

-0.95

POSITIVE LOGITS

 Surprise

1.41

opus

1.12

avia

1.06

nard

0.98

acious

0.98

hens

0.97

å¹

0.96

boro

0.94

OB

0.94

 Fool

0.93

Activations Density 0.705%

dates in the month and year format

No Comments

No Known Activations