INDEX

Explanations

function annotations and metadata in programming-related text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 41-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_41/width_16k/average_l0_113

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.41.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

 majority

-0.48

 preservation

-0.43

…?

-0.42

 beating

-0.41

 diploma

-0.40

 rarity

-0.40

 astronaut

-0.40

fat

-0.40

 regular

-0.39

POSITIVE LOGITS

*/

1.16

**/

0.91

*/

0.81

**/

0.80

*/,

0.72

*/;

0.69

0.66

 ***/

0.64

*/

0.62

**/

0.59

Activations Density 0.078%

function annotations and metadata in programming-related text

No Comments

No Known Activations