INDEX

Explanations

programming function definitions and associated code structures

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 disambiguazione

-0.68

GEBURTS

-0.67

posedge

-0.65

:✨

-0.65

Jeografia

-0.61

曖昧さ回避

-0.60

 lenker

-0.59

 Administrativna

-0.59

 snippetHide

-0.57

onViewCreated

-0.57

POSITIVE LOGITS

OUNTS

0.34

klä

0.32

 function

0.32

hogy

0.31

mentos

0.31

 Invention

0.30

 functions

0.30

Ser

0.30

ida

0.29

NSIndexPath

0.29

Activations Density 0.012%

programming function definitions and associated code structures

No Comments

No Known Activations