INDEX

Explanations

mathematical symbols and terms used in equations and proofs

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ConstraintMaker

-0.62

 utafitiHapana

-0.45

ViewChild

-0.44

+#+#

-0.43

⤹

-0.42

 snippetHide

-0.42

ftagPool

-0.41

 jsPsych

-0.40

rungsseite

-0.40

fillType

-0.39

POSITIVE LOGITS

readystatechange

0.48

Demografie

0.44

Werdegang

0.43

pah

0.41

 незавершена

0.40

Démographie

0.40

 disambiguazione

0.40

 venden

0.40

GREES

0.39

:✨

0.39

Activations Density 0.536%

mathematical symbols and terms used in equations and proofs

No Comments

No Known Activations