INDEX

Explanations

elements related to programming functions and methods

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

data binding and view models

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

InputBorder

-0.47

الحياه

-0.43

mybatisplus

-0.42

adari

-0.39

 snippetHide

-0.38

Tikang

-0.38

 cerimonia

-0.37

AndroidJUnit

-0.36

 Vikipedi

-0.36

 Mazar

-0.36

POSITIVE LOGITS



0.60

TypedDataSet

0.52

DataBind

0.47

UnusedPrivate

0.45

 Numerade

0.44

UVWXYZ

0.44

Референце

0.43

 newBuilder

0.43

vician

0.42

ronpa

0.41

Activations Density 0.019%

elements related to programming functions and methods

data binding and view models

No Comments

No Known Activations