INDEX

Explanations

terms related to programming constructs and data handling

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 12-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_12/width_16k/average_l0_130

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.12.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

 désert

-0.35

 ainfi

-0.33

 Anſ

-0.30

mainAxisSize

-0.30

 Jugendliche

-0.30

ConstraintMaker

-0.28

íslu

-0.28

 fumée

-0.28

pośred

-0.28

 trône

-0.27

POSITIVE LOGITS

MessageOf

0.71

 transférez

0.63

 وتسجيلات

0.61

 للمعارف

0.58

 bezeichneter

0.57

 noDo

0.57

Brainz

0.57

 Photocase

0.56

 समीक्षाओं

0.55

 linkovi

0.55

Activations Density 0.905%

terms related to programming constructs and data handling

No Comments

No Known Activations