INDEX

Explanations

terms related to parameters or attributes in coding, particularly focusing on minimum and maximum values

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 34-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_34/width_16k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.34.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

 estekak

-0.55

elemField

-0.52

 poveznice

-0.47

 nakalista

-0.45

blij

-0.44

 useStyles

-0.44

 Atty

-0.44

 Toxicol

-0.44

 Neurosci

-0.43

 useAuth

-0.43

POSITIVE LOGITS

uxxxx

0.63

ization

0.60

setupUi

0.58

Rectangle

0.52

autoreleasepool

0.51

ized

0.50

isation

0.49

ize

0.49

Enumeration

0.48

getLatitude

0.47

Activations Density 0.790%

terms related to parameters or attributes in coding, particularly focusing on minimum and maximum values

No Known Activations