Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 21-GEMMASCOPE-RES-16K
    6. 10372
    Prev
    Next
    INDEX
    Explanations

    details related to illicit activities, particularly smuggling and drug transactions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.21.hook_resid_post
    Hook Layer
    21
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Parcelize
    -0.49
    FromNib
    -0.49
    fjspx
    -0.48
     reje
    -0.46
     licenci
    -0.45
    ciller
    -0.45
     ſei
    -0.45
     spokesman
    -0.43
    getMenuInflater
    -0.43
     useStyles
    -0.43
    POSITIVE LOGITS
     secretly
    0.77
     secret
    0.75
    secret
    0.74
     clandestine
    0.73
     cland
    0.70
    秘密
    0.67
     secrecy
    0.66
     stealth
    0.66
    Secret
    0.65
     secreto
    0.61
    Activations Density 0.511%

    No Known Activations