INDEX

Explanations

codes and programming concepts related to functions, classes, and method calls in software development

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

programming function calls

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

 betweenstory

-0.64

Personensuche

-0.59

хьтан

-0.55

sizeCache

-0.54

Portail

-0.52

 Waray

-0.51

.*")]

-0.49

 referrerpolicy

-0.49

 wikipagina

-0.48

évaluateur

-0.48

POSITIVE LOGITS

 suel

0.38

IContainer

0.36

 jurí

0.36

 procédures

0.35

 cuna

0.34

constructor

0.34

スナー

0.34

 weisen

0.33

 function

0.33

ngdoc

0.33

Activations Density 0.081%

codes and programming concepts related to functions, classes, and method calls in software development

programming function calls

No Comments

No Known Activations