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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 122881
    Prev
    Next
    INDEX
    Explanations

    topics related to problems and challenges faced in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     <<<<<<<<<<<<<<
    -0.46
     مرئيه
    -0.39
     ProtoMessage
    -0.38
    HasAnnotation
    -0.36
     transfieras
    -0.34
    meanor
    -0.33
    yelidikan
    -0.33
     دى
    -0.32
    writerow
    -0.32
     ModelRenderer
    -0.31
    POSITIVE LOGITS
     problems
    0.89
     Problems
    0.79
    problems
    0.75
     Probleme
    0.74
     difficulties
    0.73
    Problems
    0.73
     problemas
    0.71
     Probl
    0.71
     dilemmas
    0.70
     problèmes
    0.70
    Activations Density 0.044%

    No Known Activations