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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & 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. 22-GEMMASCOPE-RES-16K
    6. 12978
    Prev
    Next
    INDEX
    Explanations

    functional groups and structures, particularly those related to lists and data processing in code

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

    Code that appends to lists

    np_acts-logits-general · gemini-2.0-flash

    list append and pop

    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
    isOk
    -0.60
     naturelles
    -0.59
     blessés
    -0.56
    EndContext
    -0.54
     diikuti
    -0.53
    oluciones
    -0.51
     OMITBAD
    -0.51
    simmon
    -0.49
     Intr
    -0.48
     défaut
    -0.47
    POSITIVE LOGITS
    push
    1.02
     push
    0.94
     pushed
    0.84
     pushes
    0.83
     pusher
    0.79
     pop
    0.77
     PUSH
    0.77
    pop
    0.76
     Push
    0.76
     pushing
    0.75
    Activations Density 0.104%

    No Known Activations