INDEX

Explanations

terms related to academic and legal contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 37-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_37/width_16k/average_l0_124

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.37.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

 vérit

-0.62

 nahilalakip

-0.61

 prêtres

-0.54

 argint

-0.52

 uLocal

-0.52

Véxase

-0.51

 boî

-0.51

 Privacidad

-0.51

 vœux

-0.51

 prochaines

-0.50

POSITIVE LOGITS

 các

0.68

 Hills

0.52

 bills

0.51

 acids

0.50

 salts

0.50

 baths

0.50

 HILLS

0.50

 ethers

0.50

 streams

0.49

oids

0.49

Activations Density 1.799%

terms related to academic and legal contexts

No Comments

No Known Activations

terms related to academic and legal contexts

No Comments

No Known Activations