INDEX

Explanations

terms related to improvement and organization

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

 мәкал

-0.63

хьтан

-0.52

 nahilalakip

-0.51

Билгалдахарш

-0.46

олові

-0.45

DockStyle

-0.43

Ӕ

-0.42

boneca

-0.42

 Signalez

-0.41

cambios

-0.41

POSITIVE LOGITS

ChildScrollView

0.42

quelize

0.41

onedDateTime

0.40

TokenNameRPAREN

0.39

BufferException

0.39

所

0.38

ble

0.38

$_"

0.38

 "..\..\..\

0.37

</thead>

0.37

Activations Density 0.287%

terms related to improvement and organization

No Comments

No Known Activations