INDEX

Explanations

references to academic resources and online scholarly content

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 13-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_13/width_65k/average_l0_75

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.13.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 متعلقه

-0.58

UnusedPrivate

-0.56

sme

-0.52

audiovisuel

-0.50

imetsu

-0.44

 ويكيپيديا

-0.44

FA

-0.44

///</

-0.44

そうで

-0.43

͡

-0.43

POSITIVE LOGITS

shields

0.70

 endregion

0.69

 gynhyrchwyd

0.65

reportWebVitals

0.65

reactstrap

0.64

ipients

0.64

expandindo

0.60

 universitario

0.60

Cyfeiriadau

0.60

にほんブログ村

0.59

Activations Density 0.007%

references to academic resources and online scholarly content

No Comments

No Known Activations