INDEX

Explanations

statements or declarations related to correctness and claims

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

academic or technical document review language and terminology.

oai_token-act-pair · claude-3-5-haiku-20241022 Triggered by @neilrathi

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_16k/average_l0_68

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

URLException

-0.57

localctx

-0.51

íncia

-0.49

 ivelany

-0.48

 pinulongan

-0.48

afficheront

-0.47

 doInBackground

-0.47

oredCriteria

-0.46

 wireType

-0.46

</thead>

-0.46

POSITIVE LOGITS

transQ

0.59

 Administrativna

0.40

參考資料

0.39

 lizenzfreie

0.39

 propOrder

0.38

``

0.36

 autorytatywna

0.36

0.35

 "...

0.35

|')

0.34

Activations Density 0.380%

statements or declarations related to correctness and claims

academic or technical document review language and terminology.

No Comments

No Known Activations