INDEX

Explanations

conjunctions and expressions of reasoning or causation

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

explaining causation

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

Erreferentziak

-0.49

TagMode

-0.48

rasing

-0.47

 ReactDOM

-0.46

Tip

-0.46

TIP

-0.46

 METRO

-0.46

Ace

-0.46

matchCondition

-0.45

tip

-0.44

POSITIVE LOGITS

 disambiguazione

0.50

 nahilalakip

0.41

UnsafeEnabled

0.41

cèse

0.41

 مشين

0.41

følge

0.40

″]

0.40

")));

0.38

 suaminya

0.38

 Genehmigung

0.37

Activations Density 0.050%

conjunctions and expressions of reasoning or causation

explaining causation

No Comments

No Known Activations