INDEX

Explanations

phrases related to the attribution and sources of information

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

attribution and origin phrases

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

 stål

-0.41

:✨

-0.39

 anſ

-0.39

 faſt

-0.39

Китай

-0.38

***********/

-0.36

cog

-0.35

straints

-0.35

 deſt

-0.35

Scy

-0.34

POSITIVE LOGITS

 AssemblyCompany

0.84

Datuak

0.66

 nahilalakip

0.55

haikusbot

0.53

tangentMode

0.52

ViewImports

0.51

jsxFileName

0.50

 courtesy

0.50

السكان

0.50

TabIndex

0.49

Activations Density 0.019%

phrases related to the attribution and sources of information

attribution and origin phrases

No Comments

No Known Activations