INDEX

Explanations

technical parameters and function definitions in code

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

code keywords and verbs

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

-0.61

 المعيارى

-0.55

edicated

-0.47

@@@@@

-0.46

tagHelperRunner

-0.45

 שוליים

-0.43

comprom

-0.42

fromnode

-0.41

 mxArray

-0.41

 للمعارف

-0.40

POSITIVE LOGITS

 Organisator

0.42

ambut

0.40

Damit

0.40

 llantas

0.40

LookAnd

0.39

 кӀош

0.39

âteau

0.38

 Савезне

0.37

 leçon

0.36

ंदीखरीदारी

0.36

Activations Density 0.040%

technical parameters and function definitions in code

code keywords and verbs

No Comments

No Known Activations