INDEX

Explanations

technical terms and mathematical symbols

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

ver

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

TagMode

-0.77

SequentialGroup

-0.68

Personendaten

-0.63

 resourceCulture

-0.55

 PeEnEo

-0.52

verifyException

-0.52

 незавершена

-0.48

@"/

-0.48

KommentareTeilen

-0.47

 يتيمه

-0.46

POSITIVE LOGITS

 correspondingly

0.49

Égypte

0.49

alyptus

0.48

 ſta

0.48

 desmotivaciones

0.47

 vPvB

0.46

 ainfi

0.45

 nakalista

0.45

himovic

0.45

 enfans

0.45

Activations Density 0.646%

technical terms and mathematical symbols

ver

No Comments

No Known Activations