INDEX

Explanations

references to code structure and data formatting

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 10-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_10/width_16k/average_l0_113

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.10.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

 afterwards

-0.37

 succeeding

-0.35

 eccl

-0.35

 subsequent

-0.34

 subsequently

-0.32

 eventually

-0.32

Leary

-0.31

Conclusão

-0.31

 <<<<<<<<<<<<<<

-0.31

High

-0.31

POSITIVE LOGITS

 snippetHide

0.78

__':

0.65

 defaultstate

0.64

posedge

0.63

AddHtmlAttribute

0.60

providedIn

0.59

)':

0.59

__':

0.59

"]=

0.57

yntaxException

0.57

Activations Density 0.663%

references to code structure and data formatting

No Comments

No Known Activations