INDEX

Explanations

inquiries or questions posed by users

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

referencing similar questions

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

tanleria

-0.49

 Einfluß

-0.49

፩

-0.47

EDEFAULT

-0.45

舺

-0.45

 beobachtet

-0.45

UnitTesting

-0.44

alsy

-0.44

yarnpkg

-0.44

 AssemblyVersion

-0.42

POSITIVE LOGITS

brainly

0.57

 question

0.50

question

0.48

 questions

0.48

 Questions

0.47

 Question

0.46

Questions

0.44

Question

0.42

questions

0.42

TagMode

0.41

Activations Density 0.003%

inquiries or questions posed by users

referencing similar questions

No Comments

No Known Activations