INDEX

Explanations

terms related to legal and warranty claims

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

Preceding adjectives

np_acts-logits-general · gemini-2.0-flash

adjective followed by noun

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 30-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_30/width_16k/average_l0_120

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.30.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

 nemlig

-0.48

 davon

-0.42

 dermed

-0.40

Nevertheless

-0.39

 anaknya

-0.39

 Đó

-0.38

Fortunately

-0.37

 zengin

-0.37

fortunately

-0.36

Otherwise

-0.36

POSITIVE LOGITS

OGND

0.59

 queſta

0.58

 ministerial

0.58

CppCodeGen

0.54

 episcopal

0.54

﹧

0.53

haikusbot

0.52

postolic

0.51

 regional

0.50

 好文分享

0.50

Activations Density 1.316%

terms related to legal and warranty claims

Preceding adjectives

adjective followed by noun

No Comments

No Known Activations