INDEX

Explanations

phrases that indicate sources or origins

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_9/width_16k/average_l0_73

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.9.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

 rospy

-0.60

GARET

-0.58

writeValue

-0.56

prefixer

-0.54

DNEY

-0.53

 ModelRenderer

-0.53

RegistryLite

-0.52

Obrázky

-0.52

didReceive

-0.52

gpg

-0.51

POSITIVE LOGITS

FROM

1.01

 FROM

0.88

 whence

0.88

 standpoint

0.84

From

0.83

from

0.82

 Whence

0.79

 kaynağından

0.78

 From

0.77

getFrom

0.76

Activations Density 0.569%

phrases that indicate sources or origins

No Comments

No Known Activations

phrases that indicate sources or origins

No Comments

No Known Activations