INDEX

Explanations

mentions of data models or related concepts in a quantitative context

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

cshanno/

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

 nationally

-0.42

互联网档案馆

-0.37

liğini

-0.35

 lawfully

-0.34

 считать

-0.34

 przekon

-0.33

 Palla

-0.33

 strengths

-0.33

 aufge

-0.33

 vectorielle

-0.33

POSITIVE LOGITS

 betweenstory

0.87

########.

0.82

 незавершена

0.73

Personendaten

0.72

تقاوى

0.72

 Numerade

0.70

RegressionTest

0.69

libft

0.68

addCriterion

0.66

WebVitals

0.63

Activations Density 0.000%

mentions of data models or related concepts in a quantitative context

cshanno/

No Comments

No Known Activations