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. 22-GEMMASCOPE-RES-16K
    6. 7945
    Prev
    Next
    INDEX
    Explanations

    responses that request help, corrections, or feedback, often in the context of technical or procedural issues

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

    Followed by conditional "if"

    np_acts-logits-general · gemini-2.0-flash

    conditional "if"

    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
    __':
    
    -0.84
    __":
    
    -0.83
    __':
    -0.80
    viewDidLoad
    -0.73
    __":
    -0.72
     propOrder
    -0.67
    providedIn
    -0.66
     mergeFrom
    -0.66
    MigrationBuilder
    -0.65
    PMailer
    -0.65
    POSITIVE LOGITS
     if
    3.85
    if
    2.94
     если
    2.69
     jika
    2.69
     If
    2.62
     nếu
    2.61
     dacă
    2.60
    If
    2.51
     hvis
    2.44
     якщо
    2.40
    Activations Density 2.419%

    No Known Activations