INDEX

Explanations

links between technical and professional terms, focusing on discussing practices and their implications

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ĸļ

-0.51

hoe

-0.47

orld

-0.42

idal

-0.42

 barn

-0.41

xual

-0.41

enn

-0.41

ordial

-0.41

usalem

-0.41

 Awakens

-0.41

POSITIVE LOGITS

Recommend

0.52

 Otherwise

0.48

 however

0.48

please

0.47

 corrective

0.47

 commercially

0.44

 depends

0.43

 inexpensive

0.42

 albeit

0.42

ogical

0.42

Activations Density 22.215%

links between technical and professional terms, focusing on discussing practices and their implications

No Comments

No Known Activations