INDEX

Explanations

references to examples within the text

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

the word example

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

Derbyniad

-0.45

 Conſ

-0.45

 automatiques

-0.45

 pemasaran

-0.44

sprzedam

-0.44

 chofe

-0.44

 juſt

-0.43

 Layanan

-0.43

 мәкал

-0.43

SBATCH

-0.42

POSITIVE LOGITS

 ExecuteAsync

0.58

 example

0.56

WithIOException

0.54

 instance

0.52

 للاسماء

0.50

 quer

0.48

 Example

0.47

 detail

0.46

 scenario

0.46

Example

0.45

Activations Density 0.047%

references to examples within the text

the word example

No Comments

No Known Activations