INDEX

Explanations

concepts related to higher values and concentrations in scientific contexts

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

higher or lower values

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

setVerticalGroup

-0.55

 فريبيس

-0.55

 nonUne

-0.53

WebElementEntity

-0.53

 Савезне

-0.52

__':

-0.51

__":

-0.50

:✨

-0.49

 ویکی‌پدیای

-0.47

 <=",

-0.47

POSITIVE LOGITS

Larger

0.64

Higher

0.58

larger

0.56

 Larger

0.54

 higher

0.54

Bigger

0.54

Smaller

0.54

 larger

0.52

higher

0.51

 Higher

0.48

Activations Density 0.122%

concepts related to higher values and concentrations in scientific contexts

higher or lower values

No Comments

No Known Activations