INDEX

Explanations

mentions of programming packages and classes

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

Code package names and import statements

np_acts-logits-general · gemini-2.0-flash

package names and code elements

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

 parezca

-0.39

 generally

-0.36

Quién

-0.35

 directly

-0.33

 greatly

-0.33

 berechtigt

-0.32

 looking

-0.31

 totally

-0.30

 dónde

-0.30

entière

-0.30

POSITIVE LOGITS

 defaultstate

0.93

OGND

0.89

########.

0.87

 CreateTagHelper

0.84

Diwedd

0.79

ViewImports

0.77

aarrggbb

0.77

rungsseite

0.77

 informée

0.75

 Administrativna

0.75

Activations Density 0.635%

mentions of programming packages and classes

Code package names and import statements

package names and code elements

No Comments

No Known Activations