INDEX

Explanations

computer programming terminology related to parameters and configuration

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Configuration

neuronpedia/gpt2-small__res_slefr-ajt/2-res_slefr-ajt

Prompts (Dashboard)

12,288 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

46,080

Data Type

torch.float32

Hook Point

blocks.2.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

iverpool

-0.51

SEA

-0.49

scrib

-0.48

Chip

-0.48

CAN

-0.47

ETF

-0.47

ãĥĺ

-0.46

VERT

-0.46

weather

-0.46

avery

-0.46

POSITIVE LOGITS

emis

0.46

eter

0.44

ouri

0.42

 downside

0.42

 advis

0.42

 downs

0.42

 unseen

0.42

 headaches

0.41

 outcome

0.41

 erosion

0.41

Activations Density 0.001%

computer programming terminology related to parameters and configuration

No Comments

No Known Activations