INDEX

Explanations

terms related to specific configurations or conditions in various contexts

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

related to specific items

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

 SwitchCompat

-0.66

 Wiktionnaire

-0.59

GOTREF

-0.55

MemoryWarning

-0.52

enterOuterAlt

-0.51

GEBURTSDATUM

-0.50

جغرافيا

-0.50

 חיצוניים

-0.49

 Signalez

-0.49

mittent

-0.48

POSITIVE LOGITS

 ſta

0.49

 Pingback

0.44

ViewImports

0.43

]}"

0.43

ollectionView

0.42

 ſtre

0.42

 ligiloj

0.41

},"

0.39



0.39

})->

0.39

Activations Density 0.041%

terms related to specific configurations or conditions in various contexts

related to specific items

No Comments

No Known Activations