INDEX

Explanations

terms related to low-level hardware operations and data structures

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

C structures, especially bit fields and MAC addresses

np_acts-logits-general · gemini-2.0-flash

code lengths and offsets

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

TypedDataSet

-0.63

 XCTest

-0.59

ValueStyle

-0.58

ltk

-0.56

Vidite

-0.56

 noDo

-0.55

Transkript

-0.53

 SDLK

-0.53

RTLR

-0.53

hyrchwyd

-0.52

POSITIVE LOGITS

 ligeramente

0.35

 slightly

0.35

 légèrement

0.35

slightly

0.35

 oyuncu

0.34

 hafif

0.34

 AssemblyVersion

0.34

put

0.34

 Slightly

0.33

scanning

0.33

Activations Density 1.885%

terms related to low-level hardware operations and data structures

C structures, especially bit fields and MAC addresses

code lengths and offsets

No Comments

No Known Activations