INDEX

Explanations

numerical values and measurements in the text

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

conversions and calculations

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

 typelib

-0.57

出版年

-0.56

">//

-0.54

帖最后由

-0.54

 numerus

-0.52

पया

-0.48

AccessorTable

-0.48

 للمعارف

-0.48

 للاسماء

-0.47

)))),

-0.46

POSITIVE LOGITS

 Berechnung

0.47

GeneratedMessage

0.42

printStackTrace

0.41

ogia

0.40

MILLIS

0.40

riwal

0.39

 Präsidenten

0.39

ſelf

0.38

getElementById

0.38

HostException

0.38

Activations Density 0.324%

numerical values and measurements in the text

conversions and calculations

No Comments

No Known Activations