INDEX

Explanations

elements of mathematical notation and expressions in a structured format

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

iconque

-0.34

adpleegd

-0.34

-0.32

SPATH

-0.32

 wzgl

-0.32

########.

-0.31

FieldNumber

-0.30

 corolla

-0.30

IntentFilter

-0.29

ngdoc

-0.29

POSITIVE LOGITS

 autorytatywna

0.54

 Taktlose

0.54

 Perſ

0.54

 насељу

0.52

 Reſ

0.51

 deſt

0.50

UnsafeEnabled

0.48

 Chwiliwch

0.47

 Inſ

0.47

ilding

0.46

Activations Density 0.043%

elements of mathematical notation and expressions in a structured format

No Comments

No Known Activations