INDEX

Explanations

terms related to denial and serious consequences or conditions

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

denied, serious, death, cutting, null, eliminate

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

 nahilalakip

-0.51

IntoConstraints

-0.51

addPreferredGap

-0.47

 Receiver

-0.45

 CanadaChoose

-0.45

ۜ

-0.45

jaman

-0.42

 Pomer

-0.42

 Dynasty

-0.41

 Bata

-0.40

POSITIVE LOGITS

 NSCoder

0.54

 PeEnEo

0.48

PerformLayout

0.41

jš

0.40

 downright

0.39

ágenes

0.39

 Infór

0.36

 esperienze

0.36

 lèvres

0.36

 onData

0.36

Activations Density 0.084%

terms related to denial and serious consequences or conditions

denied, serious, death, cutting, null, eliminate

No Comments

No Known Activations