INDEX

Explanations

communication-related terms and functions in a programming context

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

broadcast, tell, send, msg, data

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

 للاسماء

-0.79

 Normdatei

-0.73

SBATCH

-0.64

 NSCoder

-0.63

 autorytatywna

-0.60

 المعيارى

-0.59

 nakalista

-0.58

TagMode

-0.57

évaluateur

-0.57

AndroidJUnit

-0.57

POSITIVE LOGITS

clusal

0.37

enciales

0.35

 nghi

0.35

anneer

0.34

neid

0.33

局

0.33

ómica

0.32

 Combined

0.32

 aank

0.32

ご了承ください

0.32

Activations Density 0.005%

communication-related terms and functions in a programming context

broadcast, tell, send, msg, data

No Comments

No Known Activations