INDEX

Explanations

elements of programming or code-related commands and structures

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

code snippets and keywords

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

 صوتيه

-0.62

AddTagHelper

-0.49

rungsseite

-0.47

UnknownFields

-0.47

 newItem

-0.41

bootstrapcdn

-0.41

 continuant

-0.41

:✨

-0.40

 restes

-0.39

 remise

-0.39

POSITIVE LOGITS

IndentedString

0.50

帖最后由

0.43

msgTypes

0.40

 Holtz

0.37

 Scal

0.37

Grüsse

0.36

WebControls

0.35

ContentAlignment

0.35

 geprek

0.35

 actionPerformed

0.35

Activations Density 0.011%

elements of programming or code-related commands and structures

code snippets and keywords

No Comments

No Known Activations