INDEX

Explanations

elements related to transactional or procedural information

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

legal, policy, or technical documents

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

TagMode

-0.44

 Roskov

-0.40

 disponibilités

-0.40

rshire

-0.38

#+#

-0.37

@"/

-0.36

himovic

-0.35

 PeEnEo

-0.34

 wireType

-0.33

 SwitchCompat

-0.33

POSITIVE LOGITS

ViewFeatures

0.53

ValueStyle

0.48

styleUrls

0.47

 <<<<<<<<<<<<<<

0.44

 betweenstory

0.44

 ujednoznacz

0.44

pushFollow

0.44

 Kebijakan

0.43

 jadx

0.42

 Absorption

0.41

Activations Density 0.025%

elements related to transactional or procedural information

legal, policy, or technical documents

No Comments

No Known Activations