INDEX

Explanations

terms related to network systems and data communication

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

 становника

-0.44

žiai

-0.31

close

-0.31

 Vieira

-0.31

Harry

-0.29

LOADED

-0.29

 ولو

-0.28

tschafts

-0.28

arbeitung

-0.27

BR

-0.26

POSITIVE LOGITS

 <=",

0.67

 contextLoads

0.62

 <>",

0.60

erçe

0.59

 jScrollPane

0.58

 dAtA

0.57

 dashboards

0.56

 gateways

0.54

ErrUnexpectedEOF

0.54

 fieldNum

0.54

Activations Density 0.025%

terms related to network systems and data communication

No Comments

No Known Activations