INDEX

Explanations

terms and notation related to mathematical functions or concepts

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

 CreateTagHelper

-0.56

MessageTagHelper

-0.54

 Biôgrafia

-0.52

 SqlCommand

-0.52

xase

-0.50

 informée

-0.48

 EClass

-0.48

PerformLayout

-0.46

Portale

-0.46

BASEPATH

-0.46

POSITIVE LOGITS

 translate

0.40

 zählen

0.37

 counted

0.37

 translates

0.36

 count

0.36

 Herkunft

0.36

 eigenes

0.35

0.34

 Vorge

0.34

Translate

0.33

Activations Density 0.043%

terms and notation related to mathematical functions or concepts

No Comments

No Known Activations