INDEX

Explanations

elements related to unique languages and their usage

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

Verbs followed by prepositions

np_acts-logits-general · gemini-2.0-flash

past action or production

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 24-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_24/width_16k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.24.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

ValueStyle

-0.54

 Succ

-0.48

 HttpNotFound

-0.48

 <<<<<<<<<<<<<<

-0.47

 jälkeen

-0.47

trui

-0.45

 Beasley

-0.45

 preventiva

-0.43

FunctionFlags

-0.43

 kehilangan

-0.43

POSITIVE LOGITS

 propOrder

0.60

rungsseite

0.40

 насељу

0.36

 дописавши

0.35

 betweenstory

0.35

 exist

0.34

gypti

0.34

不下

0.33

 currently

0.32

 &___

0.32

Activations Density 0.629%

elements related to unique languages and their usage

Verbs followed by prepositions

past action or production

No Comments

No Known Activations