INDEX

Explanations

mathematical variables and equations within a programming context

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

vector lengths and angles

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

🤶

-0.40

rasil

-0.39

eroon

-0.37

 Aula

-0.36

 мәкал

-0.36

 ujednoznacz

-0.36

esgue

-0.35

MockMvc

-0.35

 Econ

-0.34

 suspens

-0.34

POSITIVE LOGITS

yntaxException

0.63

 stazione

0.47

 station

0.46

 point

0.46

GTCX

0.45

 lusso

0.43

Kjelder

0.42

 nearby

0.41

 paziente

0.41

Autoritní

0.41

Activations Density 0.094%

mathematical variables and equations within a programming context

vector lengths and angles

No Comments

No Known Activations