INDEX

Explanations

programming statements and constructs typically used in code

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 12-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_12/width_16k/average_l0_130

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.12.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

########.

-0.65

OGND

-0.59

 مشين

-0.59

ſelf

-0.57

новништво

-0.56

clerView

-0.56

fieur

-0.56

Erreferentziak

-0.54

 snippetHide

-0.54

principalColumn

-0.53

POSITIVE LOGITS

 leão

0.33

 dima

0.32

cos

0.32

sin

0.31

 żyw

0.31

Nec

0.30

ATIVO

0.30

Nec

0.30

intra

0.30

ness

0.29

Activations Density 0.487%

programming statements and constructs typically used in code

No Comments

No Known Activations

programming statements and constructs typically used in code

No Comments

No Known Activations