INDEX

Explanations

brand names and product identifiers

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

`axone` or `DMD patients`

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

WriteBarrier

-0.44

䛗

-0.40

ritch

-0.34

__*/

-0.34

 milit

-0.34

chic

-0.34

즘

-0.33

 difficult

-0.33

alho

-0.33

醐

-0.33

POSITIVE LOGITS

 betweenstory

0.73

WebElementEntity

0.63

AISSEE

0.57

 Italijanski

0.54

 intptr

0.54

########.

0.53

InstrumentedTest

0.52

 ujednoznacz

0.52

transQ

0.52

 isNameExpr

0.50

Activations Density 0.017%

brand names and product identifiers

`axone` or `DMD patients`

No Comments

No Known Activations