INDEX

Explanations

expressions indicating reasoning and justification

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

giving reasons before a conclusion

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

:✨

-0.56

Autoritní

-0.56

MigrationBuilder

-0.54

ंदीखरीदारी

-0.53

tagext

-0.52

principalColumn

-0.51

WillAppear

-0.50

səhifə

-0.49

 صوتيه

-0.48

 CanadaChoose

-0.46

POSITIVE LOGITS

gemaakt

0.41

 TextAlign

0.37

leiding

0.36

transQ

0.36

 controversy

0.35

的一切

0.35

DialogInterface

0.34

事です

0.33

Abp

0.33

 GenerationType

0.33

Activations Density 0.113%

expressions indicating reasoning and justification

giving reasons before a conclusion

No Comments

No Known Activations