INDEX

Explanations

comparisons of quantities or numbers

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs24576-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 masses

-0.62

onom

-0.62

ļéĨĴ

-0.61

onomy

-0.59

 Newsletter

-0.58

 photographs

-0.57

([

-0.57

rive

-0.57

olitical

-0.56

ug

-0.56

POSITIVE LOGITS

 equivalent

0.79

berus

0.75

lihood

0.75

 compared

0.73

ilion

0.68

pees

0.68

 surpassed

0.66

actly

0.63

ceivable

0.63

 Ortiz

0.62

Activations Density 0.233%

comparisons of quantities or numbers

No Comments

No Known Activations