INDEX

Explanations

programming language names and their associated terminology

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

swift and protobuf

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

tahui

-0.45

AndEndTag

-0.43

niająca

-0.41

principalTable

-0.39

RenderAtEndOf

-0.37

 resourceCulture

-0.37

tagHelperRunner

-0.36

 карьера

-0.35

 незавершена

-0.35

 הע

-0.35

POSITIVE LOGITS

})*/

0.57

()}>

0.56

 femininos

0.52

 modernos

0.51

)"),

0.51

 modernas

0.50

")]

0.49

 Bewußt

0.46

 AspNetCore

0.45

RTSN

0.45

Activations Density 0.030%

programming language names and their associated terminology

swift and protobuf

No Comments

No Known Activations