INDEX

Explanations

terms related to inquiry and understanding in a procedural or evaluative context

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

 ویکی‌پدیا

-0.49

ConstraintMaker

-0.48

typeorm

-0.47

الحياه

-0.46

帖最后由

-0.46

 كومونز

-0.45

 valentín

-0.45

archiviato

-0.44

XmlAccessType

-0.42

 royaume

-0.40

POSITIVE LOGITS

CreateInfo

0.40

luor

0.40

:✨

0.40

tigten

0.39

Pub

0.39

 године

0.39

ɚ

0.38

errHandler

0.38

<>();

0.37

INCT

0.37

Activations Density 0.096%

terms related to inquiry and understanding in a procedural or evaluative context

No Comments

No Known Activations