INDEX

Explanations

technical specifications and characteristics related to training devices and systems

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

suited for specific applications

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

featureID

-0.51

 surla

-0.49

Demografía

-0.46

Tembelea

-0.46

 '\\;'

-0.46

évaluateur

-0.46

 المعيارى

-0.45

Personendaten

-0.44

AddTagHelper

-0.43

🇶

-0.42

POSITIVE LOGITS

 requirements

0.44

 preferred

0.39

izr

0.38

 preference

0.38

 desired

0.38

 Philipp

0.37

 Curitiba

0.36

ffilm

0.36

 REQUIREMENTS

0.36

 tailor

0.35

Activations Density 0.029%

technical specifications and characteristics related to training devices and systems

suited for specific applications

No Comments

No Known Activations