INDEX

Explanations

elements related to measurements and numerical data

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 36-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_36/width_16k/average_l0_120

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.36.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

الإنجليزية

-0.69

Diweddarwch

-0.61

AsUp

-0.58



-0.56

#+#

-0.55

__':

-0.54

__':

-0.54

 propOrder

-0.54

makeConstraints

-0.53

RTCF

-0.52

POSITIVE LOGITS

참고

0.47

 souverain

0.42

rigin

0.41

ZERO

0.40

 становника

0.40

 Gewalt

0.39

zero

0.39

Cobalt

0.38

 régal

0.37

Activations Density 0.839%

elements related to measurements and numerical data

No Comments

No Known Activations