INDEX

Explanations

content related to conservative news and updates

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

questions about personal 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

SequentialGroup

-0.54

HostException

-0.51

 resourceCulture

-0.50

InjectAttribute

-0.47

atguigu

-0.46

principalTable

-0.46

recated

-0.45

rmtree

-0.45

BagConstraints

-0.45

openConnection

-0.43

POSITIVE LOGITS

 ویکی‌پدی

0.38

⟬

0.35

 policiers

0.35

Filmographie

0.34

 scolaires

0.34

 tenerife

0.33

 financières

0.33

 urgente

0.33

 <=",

0.32

 crucero

0.32

Activations Density 0.039%

content related to conservative news and updates

questions about personal information

No Comments

No Known Activations