INDEX

Explanations

numerical data, particularly in a structured format like lists or sequences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 10-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_10/width_16k/average_l0_77

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.10.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

+#+#

-0.89

contentLoaded

-0.89

 مشين

-0.83

 للمعارف

-0.82

SequentialGroup

-0.81

 useAppContext

-0.79

.*")]

-0.76

__':

-0.76

 tartalomajánló

-0.74

InjectAttribute

-0.73

POSITIVE LOGITS

0.48

ibouti

0.45

0.44

0.43

0.42

wack

0.41

0.40

Activations Density 1.012%

numerical data, particularly in a structured format like lists or sequences

No Comments

No Known Activations