INDEX

Explanations

code snippets related to programming languages and functions

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

ValueStyle

-0.61

 getCode

-0.57

 queſta

-0.55

 bookId

-0.54

$_(

-0.54

$_"

-0.53

)++;

-0.51

 itemId

-0.50

 mData

-0.50

 userType

-0.50

POSITIVE LOGITS

 again

0.40

 replacement

0.40

 kembali

0.39

 reset

0.38

reset

0.37

 linkovi

0.36

 tekrar

0.36

 novamente

0.35

 вновь

0.35

 yeniden

0.34

Activations Density 0.521%

code snippets related to programming languages and functions

No Comments

No Known Activations