INDEX

Explanations

technical terms related to computing and technology

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

optimization and detention

np_acts-logits-general · gemini-2.5-flash-lite

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

rungsseite

-0.75

 للمعارف

-0.60

//

-0.57

OGND

-0.55

 ExecuteAsync

-0.54

AutoScaleMode

-0.51

 يتيمه

-0.51

出版年

-0.50

 disambiguazione

-0.49

webElementXpaths

-0.49

POSITIVE LOGITS

 notwithstanding

0.44

怎么办

0.43

Carcinogenicity

0.41

 azonban

0.40

 meanwhile

0.40

vs

0.40

Conclusión

0.39

怎麼辦

0.39

 however

0.39

Dicas

0.37

Activations Density 0.371%

technical terms related to computing and technology

optimization and detention

No Comments

No Known Activations