INDEX

Explanations

numbers and percentages related to data points or statistics

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

session . patch . award . prize

np_acts-logits-general · gemini-2.5-flash-lite

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 resourceCulture

-0.58

EndInit

-0.55

 برانيه

-0.55

Diweddarwch

-0.54

mybatisplus

-0.52

 nakalista

-0.49

BeginContext

-0.49

__(/*!

-0.45

RegistryLite

-0.45

цездатний

-0.43

POSITIVE LOGITS

 jsPsych

0.44

spans

0.39

Voci

0.38

ölkerung

0.37

헌

0.37

 ElevatedButton

0.36

isesi

0.36

kově

0.36

 Large

0.36

neurons

0.36

Activations Density 0.377%

numbers and percentages related to data points or statistics

session . patch . award . prize

No Comments

No Known Activations