INDEX

Explanations

legal and regulatory terms related to liability and responsibility

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

qualities and states

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

SBATCH

-0.50

GEBURTSDATUM

-0.47

nodoc

-0.45

 endblock

-0.43

plenty

-0.43

 sobra

-0.42

 gekomen

-0.42

Plenty

-0.41

 huelga

-0.41

šech

-0.40

POSITIVE LOGITS

Personensuche

0.46

 nakalista

0.45

Bakgrunnsstoff

0.43

delwed

0.39

 lenker

0.39

Попис

0.38

 accessibility

0.38

 ability

0.37

Становништво

0.35

intenant

0.34

Activations Density 0.272%

legal and regulatory terms related to liability and responsibility

qualities and states

No Comments

No Known Activations