INDEX

Explanations

elements of programming code, particularly related to initialization and expression syntax

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

code snippets around values

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

AndEndTag

-0.81

 تكبرها

-0.79

 للمعارف

-0.74

 ब्रेकडाउन

-0.73

ロウィン

-0.72

 EconPapers

-0.71

 AssemblyCulture

-0.69

endphp

-0.68

出版年

-0.68

transQ

-0.66

POSITIVE LOGITS

setVerticalGroup

0.46

DockStyle

0.34

oneofs

0.34

să

0.29

té

0.28

0.27

usammen

0.27

 Gesam

0.27

ledning

0.26

føl

0.25

Activations Density 0.046%

elements of programming code, particularly related to initialization and expression syntax

code snippets around values

No Comments

No Known Activations