INDEX

Explanations

references to previous cases and the structure of findings related to studies

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

repetition or sequence

np_acts-logits-general · gemini-2.0-flash

temporal sequence or repetition

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

aarrggbb

-0.68

 للاسماء

-0.56

rungsseite

-0.54

Fordítás

-0.49

 betweenstory

-0.49

astéro

-0.46

 createSprite

-0.45

ellees

-0.45

 мәкал

-0.45

niſſe

-0.44

POSITIVE LOGITS

 repeatedly

0.47

 progressively

0.45

 repeated

0.44

 continually

0.43

 twice

0.42

 mehrm

0.42

 Contin

0.41

 прогре

0.41

 continuously

0.41

 liên

0.41

Activations Density 1.047%

references to previous cases and the structure of findings related to studies

repetition or sequence

temporal sequence or repetition

No Comments

No Known Activations