INDEX

Explanations

key phrases related to descriptions of entities and their functions or characteristics

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

connectors or states of being

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

 kasarigan

-0.64

 propOrder

-0.64

Autoritní

-0.61

webElementXpaths

-0.59

 kohdetta

-0.57

FunctionFlags

-0.56

:✨

-0.56

 Италијани

-0.55

oa̍t

-0.54

MessageOf

-0.54

POSITIVE LOGITS

 also

0.47

 ayrıca

0.41

 विश्वसनीयता

0.39

 additionally

0.38

 Vaters

0.38

 همچنین

0.36

 satunya

0.36

findpost

0.34

Initializable

0.34

ExtendWith

0.34

Activations Density 0.017%

key phrases related to descriptions of entities and their functions or characteristics

connectors or states of being

No Comments

No Known Activations