INDEX

Explanations

function calls related to memory management and cleanup

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

rophobic

-0.63

$_['

-0.61

XmlAccessType

-0.59

racite

-0.59

bootstrapcdn

-0.58

transQ

-0.57

<unused47>

-0.56

<unused68>

-0.56

<unused42>

-0.56

<unused28>

-0.56

POSITIVE LOGITS

EndContext

0.41

 forever

0.33

return

0.32

Hentet

0.32

EndInit

0.32

sp

0.32

 aDecoder

0.30

 judiciales

0.30

 completely

0.29

CreateIndex

0.28

Activations Density 0.013%

function calls related to memory management and cleanup

No Comments

No Known Activations