INDEX

Explanations

references to product availability and notifications

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

followed by citations/references

np_acts-logits-general · gemini-2.0-flash

secret, tape, attempts, forcibly

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-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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

__*/

-0.36

누

-0.35

uttgart

-0.32

 material

-0.30

 مشين

-0.30

бурга

-0.30

 Bouch

-0.30

뀜

-0.29

mac

-0.28

掏

-0.28

POSITIVE LOGITS

ChildScrollView

0.61

 الرياضيه

0.60

 compensating

0.58

compensation

0.55

 للمعارف

0.54

 compensate

0.54

 compensation

0.52

 unknownFields

0.51

 informée

0.50

 فريبيس

0.50

Activations Density 0.778%

references to product availability and notifications

followed by citations/references

secret, tape, attempts, forcibly

No Comments

No Known Activations