INDEX

Explanations

additional qualifiers or modifiers indicating extra or supplementary information

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

additional or supplemental information

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

 NSCoder

-0.52

Portale

-0.44

 veľa

-0.42

Glej

-0.38

 tartalomajánló

-0.37

 Dress

-0.36

 intStringLen

-0.36

ARROLL

-0.36

 leiding

-0.35

aclk

-0.34

POSITIVE LOGITS

帖最后由

0.51

Бахар

0.51

Халык

0.49

 Chwiliwch

0.48

$_['

0.48

⟭

0.47

fortawesome

0.47

новништво

0.47

fjspx

0.47

原始内容存档于

0.44

Activations Density 0.071%

additional qualifiers or modifiers indicating extra or supplementary information

additional or supplemental information

No Comments

No Known Activations