INDEX

Explanations

references to nested structures and their attributes in programming contexts

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

code entities and types

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

 Kette

-0.27

 katkı

-0.24

アナ

-0.23

 Mannschaft

-0.23

 comenz

-0.23

数百

-0.23

 suun

-0.23

 Chemical

-0.23

myra

-0.22

 Feld

-0.22

POSITIVE LOGITS

UnusedPrivate

0.65

dafx

0.64

 queſta

0.63

 betweenstory

0.62

 chi̍t

0.61

writeFieldEnd

0.60

 kaynağından

0.60

MemoryWarning

0.60

#+#

0.59

хьтан

0.59

Activations Density 0.201%

references to nested structures and their attributes in programming contexts

code entities and types

No Comments

No Known Activations