INDEX

Explanations

statistical references and performance metrics related to algorithms

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

metric scores and reliability

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

 informée

-0.90

 noDo

-0.67

сылкі

-0.56

 EconPapers

-0.52

 autorytatywna

-0.48

цездатний

-0.45

ItemBackground

-0.45

 ffilmiau

-0.44

 kasarigan

-0.43

onAttach

-0.42

POSITIVE LOGITS

 accuracy

0.51

 helft

0.46

 Accuracy

0.42

 consistency

0.42

 optimum

0.41

均

0.41

 smooth

0.41

 efficiency

0.41

RTCK

0.40

 reliability

0.40

Activations Density 0.132%

statistical references and performance metrics related to algorithms

metric scores and reliability

No Comments

No Known Activations