INDEX

Explanations

sections of text related to data structures or code components

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

android ui layout xml

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

 Einfluß

-0.56

 esclavos

-0.55

 '\\;'

-0.55

RegressionTest

-0.53

reportWebVitals

-0.52

 feroit

-0.52

 avoient

-0.48

 ainfi

-0.48

 étoit

-0.47

AddTagHelper

-0.47

POSITIVE LOGITS

 Scooby

0.55

 Davos

0.53

 Warp

0.53

 Baba

0.52

相手に

0.49

 Mannheim

0.49

 accessibility

0.48

 Daredevil

0.48

 patron

0.48

 القدر

0.48

Activations Density 0.004%

sections of text related to data structures or code components

android ui layout xml

No Comments

No Known Activations