INDEX

Explanations

terms related to origin and network hosts

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

source or origin

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

AndEndTag

-0.82

rungsseite

-0.73

 Infórmanos

-0.68

 kaarangay

-0.68

Efq

-0.65

exitRule

-0.65

InSection

-0.61

Autoritní

-0.60

Clik

-0.58

Hochspringen

-0.58

POSITIVE LOGITS

CrossOrigin

0.58

Origin

0.54

 origin

0.47

origin

0.46

ORIGIN

0.46

 Origin

0.43

 origem

0.41

 origins

0.40

 konsumsi

0.38

 origen

0.37

Activations Density 0.001%

terms related to origin and network hosts

source or origin

No Comments

No Known Activations