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-2-2B
    4. Residual Stream - 16k
    5. 24-GEMMASCOPE-RES-16K
    6. 2371
    Prev
    Next
    INDEX
    Explanations

    code elements related to threading and asynchronous processes

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

    Native code, JNI, or FFI calls

    np_acts-logits-general · gemini-2.0-flash

    native code identifiers

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 24-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_24/width_16k/average_l0_73
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.24.hook_resid_post
    Hook Layer
    24
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ="@+
    -0.61
    iddhar
    -0.59
    omycin
    -0.57
    UrlResolution
    -0.57
    ConstraintMaker
    -0.53
    ftagPool
    -0.52
    Legături
    -0.48
     gql
    -0.47
    ividual
    -0.45
    SizeMode
    -0.44
    POSITIVE LOGITS
     native
    1.08
    native
    0.93
    Native
    0.91
     Native
    0.91
     Natives
    0.88
     unsafe
    0.85
     natives
    0.84
     nativo
    0.81
     nativa
    0.77
    unsafe
    0.75
    Activations Density 0.768%

    No Known Activations