INDEX

Explanations

logical connections and implications in mathematical or technical contexts

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

implies or hence

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

 Monfieur

-0.74

 ſche

-0.62

 Anſ

-0.61

 ſtate

-0.61

 ſtand

-0.59

ISupport

-0.57

 Verſ

-0.57

 Geſ

-0.56

🇶

-0.55

TypedDataSet

-0.55

POSITIVE LOGITS

 thus

0.38

 للاسماء

0.37

loem

0.34

thus

0.33

ティア

0.31

__((

0.31

øj

0.30

 dermed

0.30

 means

0.30

 Hall

0.29

Activations Density 0.242%

logical connections and implications in mathematical or technical contexts

implies or hence

No Comments

No Known Activations