© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 28-GEMMASCOPE-RES-16K
    6. 7566
    Prev
    Next
    INDEX
    Explanations

    phrases related to prevention and restrictions on actions or events

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    Followed by "from"

    np_acts-logits-general · gemini-2.0-flash

    preventing something from occurring

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 28-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_28/width_16k/average_l0_119
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.28.hook_resid_post
    Hook Layer
    28
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    🔕
    -0.33
     ży
    -0.27
     command
    -0.26
     lentement
    -0.26
     No
    -0.26
     Dig
    -0.25
     médié
    -0.25
    positiva
    -0.25
    privile
    -0.25
     Flat
    -0.25
    POSITIVE LOGITS
    IntoConstraints
    0.85
     صوتيه
    0.67
     betweenstory
    0.66
     CreateTagHelper
    0.65
     from
    0.65
    MessageOf
    0.63
    Prevent
    0.62
    RTGC
    0.61
     khỏi
    0.60
     snippetHide
    0.60
    Activations Density 0.283%

    No Known Activations