INDEX

Explanations

mentions of programming concepts and keywords

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

new language features

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

awtextra

-0.58

ſelves

-0.55

 enfans

-0.47

Datuak

-0.45

 pleaſure

-0.45

ſelf

-0.43

NameInMap

-0.42

 houſe

-0.42

MemoryWarning

-0.41

iſter

-0.41

POSITIVE LOGITS

.*")]

0.52

/*---

0.48

useNewUrlParser

0.39

 semp

0.38

BackPressed

0.38

thansa

0.37

ظ

0.37

 implemented

0.36

DataAnnotations

0.36

 insuffisamment

0.36

Activations Density 0.074%

mentions of programming concepts and keywords

new language features

No Comments

No Known Activations