INDEX

Explanations

connections between symptoms and factors or conditions

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

connecting additional information with also

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

出版年

-0.58

 pinulongan

-0.47

lably

-0.46

ItemLayout

-0.45

tableFuture

-0.45

 indiana

-0.44

Sklici

-0.44

 ProtoMessage

-0.43

ORPG

-0.43

 officielles

-0.43

POSITIVE LOGITS

 nakalista

0.42

occasion

0.41

WithIOException

0.40

 also

0.38

 Administrativna

0.38

 other

0.36

ècie

0.36

CloseOperation

0.36

 altri

0.35

scope

0.35

Activations Density 0.108%

connections between symptoms and factors or conditions

connecting additional information with also

No Comments

No Known Activations