INDEX

Explanations

dialogue and questions within a narrative context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

TargetException

-0.53

 tartalomajánló

-0.49

Clik

-0.48

MethodManager

-0.47

transQ

-0.46

析

-0.43

Hozzáférés

-0.42

 estekak

-0.42

Transkript

-0.41

 Specialization

-0.41

POSITIVE LOGITS

 Administrativna

0.46

 ब्रेकडाउन

0.40

nobyl

0.39

хьтан

0.38

oredCriteria

0.37

addCriterion

0.37

 ceva

0.36

 onBind

0.36

jsxFileName

0.36

 baloncesto

0.36

Activations Density 0.073%

dialogue and questions within a narrative context

No Comments

No Known Activations