INDEX

Explanations

key programming constructs and definitions in code snippets

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

code or math variable name

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

 propOrder

-0.62

UnusedPrivate

-0.61

disposing

-0.57

SharedCtor

-0.57

 <>",

-0.56

StoryboardSegue

-0.55

featureID

-0.52

 AspNetCore

-0.51

ISupport

-0.49

addPreferredGap

-0.49

POSITIVE LOGITS

 autorytatywna

0.40

AndEndTag

0.39

同

0.38

 flatly

0.35

}`).

0.33

drucks

0.32

fasst

0.32

"}")

0.32

 cons

0.31

搭

0.31

Activations Density 0.447%

key programming constructs and definitions in code snippets

code or math variable name

No Comments

No Known Activations