INDEX

Explanations

references to keys and key-value pairs in programming and data structures

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

key and head in code

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

脚注の使い方

-0.79

+#+#

-0.67

tangentMode

-0.61

 protoimpl

-0.60

TagMode

-0.60

tagHelperRunner

-0.56

langkah

-0.55

 disambiguazione

-0.54

UnsafeEnabled

-0.53

 kasarigan

-0.51

POSITIVE LOGITS

Embeddable

0.40

 Praze

0.40

 Theſe

0.39

سياسي

0.35

PathVariable

0.34

 alfabe

0.32

ListItemIcon

0.32

 demandas

0.31

keySet

0.29

sidemargin

0.29

Activations Density 0.018%

references to keys and key-value pairs in programming and data structures

key and head in code

No Comments

No Known Activations