INDEX

Explanations

sections of code that indicate class and method definitions within programming languages

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

access modifiers

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

UrlResolution

-0.66

 تضيفلها

-0.62

 nakalista

-0.56

-------------</

-0.54

новниш

-0.54

发表于

-0.52

 esternos

-0.52

ğraf

-0.51

 Мексичка

-0.50

 Wikiseite

-0.49

POSITIVE LOGITS

 initComponents

0.64

 Paglinawan

0.47

 Salford

0.46

Slf

0.45

patho

0.41

Linden

0.41

uct

0.41

matis

0.40

karen

0.40

 Song

0.40

Activations Density 0.005%

sections of code that indicate class and method definitions within programming languages

access modifiers

No Comments

No Known Activations