INDEX

Explanations

sections acknowledging contributions and support from others in research

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

…

-0.48

you

-0.43

-0.42

 llegará

-0.41

ystemet

-0.41

例

-0.40

 lets

-0.40

上

-0.40

宰

-0.40

POSITIVE LOGITS

Autoritní

0.83

 authors

0.79

 Authors

0.76

 lenker

0.74

 NACIONAL

0.69

 AssemblyCulture

0.69

 getCustomer

0.69

 financière

0.68

 gratefully

0.66

 manuscript

0.66

Activations Density 0.744%

sections acknowledging contributions and support from others in research

No Comments

No Known Activations