INDEX

Explanations

phrases related to security and protection

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

security and protection

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

 Infórmanos

-0.51

 '\\;'

-0.48

abbaye

-0.47

цездатний

-0.46

 ujednoznacz

-0.44

 painters

-0.43

 HasFactory

-0.43

+#+

-0.43

berdayakan

-0.42

 parasites

-0.42

POSITIVE LOGITS

 resist

0.43

resist

0.43

 resistant

0.40

resistant

0.39

 RESIST

0.38

 resists

0.37

StoryboardSegue

0.37

 resistencia

0.36

 resistance

0.36

 locked

0.36

Activations Density 0.126%

phrases related to security and protection

security and protection

No Comments

No Known Activations