© 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-2-2B
    4. Residual Stream - 16k
    5. 22-GEMMASCOPE-RES-16K
    6. 8903
    Prev
    Next
    INDEX
    Explanations

    opening curly braces in code

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

    Code snippets with identifiers/variables

    np_acts-logits-general · gemini-2.0-flash

    code syntax constructs

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.hook_resid_post
    Hook Layer
    22
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     fer
    -0.52
     chủ
    -0.49
     cu
    -0.48
     brand
    -0.48
     inter
    -0.47
    -
    -0.47
     f
    -0.46
     viv
    -0.46
    RetentionPolicy
    -0.46
    mani
    -0.45
    POSITIVE LOGITS
     Савезне
    0.95
    TagMode
    0.91
    HostException
    0.90
     Exacts
    0.86
    paravant
    0.80
    AccessorTable
    0.78
    ThroughAttribute
    0.78
     متعلقه
    0.77
    Personensuche
    0.73
    }');
    0.73
    Activations Density 0.029%

    No Known Activations