INDEX

Explanations

terms related to experimental design and optimization methods

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

experimental design and factorial analysis

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

jsxFileName

-0.43

 يتيمه

-0.39

SizeMode

-0.38

 дописавши

-0.38

 кли

-0.37

GIVEREF

-0.35

sco

-0.35

dogs

-0.35

 cachorros

-0.34

 localObject

-0.33

POSITIVE LOGITS

AndEndTag

0.52

tangentMode

0.50

 productivo

0.48

 Lider

0.47

 randomized

0.43

 propOrder

0.42

 bancada

0.42

StructEnd

0.42

bildēt

0.41

 experimento

0.41

Activations Density 0.018%

terms related to experimental design and optimization methods

experimental design and factorial analysis

No Comments

No Known Activations