INDEX

Explanations

programming constructs and functions related to object manipulation

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

underscore usage 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

UpInside

-0.66

 EconPapers

-0.61

providedIn

-0.56

 CanadaChoose

-0.55

 <>",

-0.51

Portail

-0.51

IBOutlet

-0.50

 ligiloj

-0.50

sonaro

-0.48

LEncoder

-0.48

POSITIVE LOGITS

lodash

0.81

underscore

0.72

(_.

0.59

libft

0.51

 Administrativna

0.48

 AutoMapper

0.48

########.

0.47

grunt

0.41

 methodologies

0.41

 hierbas

0.41

Activations Density 0.005%

programming constructs and functions related to object manipulation

underscore usage in code

No Comments

No Known Activations