INDEX

Explanations

mathematical or statistical terms indicating division or order

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-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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

évaluateur

-0.65

UnusedPrivate

-0.63

RegressionTest

-0.61

principalColumn

-0.56

SequentialGroup

-0.56

 betweenstory

-0.56

dirkan

-0.55

Sucesor

-0.52

MemoryWarning

-0.52

はじめに

-0.50

POSITIVE LOGITS

Билгалдахарш

0.39

󠁣

0.38

 مرئيه

0.38

 saites

0.36

ьажоргаш

0.34

 gynhyrchwyd

0.34

 мәкалә

0.34

RTLI

0.32

tut

0.31

🦰

0.31

Activations Density 0.107%

mathematical or statistical terms indicating division or order

No Comments

No Known Activations