INDEX

Explanations

references to specific items or properties associated with databases or items in a structured format

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 المعيارى

-0.48

 BoxFit

-0.37

 Quart

-0.37

WebElementEntity

-0.37

 guardi

-0.36

 Cyclo

-0.35

絶

-0.34

 صوتيه

-0.34

 sell

-0.34

 exterior

-0.34

POSITIVE LOGITS

httphttps

0.56

scriptcase

0.49

 utafitiHapana

0.49

tualmente

0.48

 betweenstory

0.47

 enfans

0.47

umumkan

0.47

 ujednoznacz

0.46

 itinéraires

0.46

exitRule

0.45

Activations Density 1.826%

references to specific items or properties associated with databases or items in a structured format

No Comments

No Known Activations