INDEX

Explanations

textual elements indicating statistical or numerical information

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

What is

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

хьтан

-0.51

 Roskov

-0.49

thâu

-0.42

 tartalomajánló

-0.42

 Chbosky

-0.40

 Италијани

-0.40

Capa

-0.40

 Garrison

-0.40

 الرياضيه

-0.40

ittarius

-0.40

POSITIVE LOGITS

 notícia

0.42

他にも

0.41

MLLoader

0.40

áld

0.40

éndolo

0.39

ValueGeneration

0.39

others

0.38

 féminin

0.38

 flavours

0.38

同じく

0.37

Activations Density 0.014%

textual elements indicating statistical or numerical information

What is

No Comments

No Known Activations

textual elements indicating statistical or numerical information

What is

No Comments

No Known Activations