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. 24-GEMMASCOPE-RES-16K
    6. 863
    Prev
    Next
    INDEX
    Explanations

    references to new concepts or experiences that involve change and adaptation

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

    Followed by "for"

    np_acts-logits-general · gemini-2.0-flash

    newness or unfamiliar experience

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 24-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_24/width_16k/average_l0_114
    Prompts (Dashboard)
    24,576 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
    ionista
    -0.43
     fidé
    -0.40
    setVerticalGroup
    -0.38
    ChildScrollView
    -0.38
    ómicas
    -0.37
     canina
    -0.36
    haustible
    -0.35
    værende
    -0.35
    risi
    -0.35
     Rossa
    -0.35
    POSITIVE LOGITS
     unfamiliar
    1.06
     unacc
    0.81
     ungew
    0.62
     accustomed
    0.61
    陌
    0.58
     inexperienced
    0.56
    慣れて
    0.55
    慣
    0.55
    初めての
    0.55
    ustomed
    0.55
    Activations Density 0.365%

    No Known Activations