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-2-2B
    4. Residual Stream - 65k
    5. 17-GEMMASCOPE-RES-65K
    6. 28533
    Prev
    Next
    INDEX
    Explanations

    punctuation marks and their roles within sentences

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 17-gemmascope-res-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_17/width_65k/average_l0_125
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    Data Type
    float32
    Hook Name
    blocks.17.hook_resid_post
    Hook Layer
    17
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    DockStyle
    -0.83
     resourceCulture
    -0.77
     getItemId
    -0.71
    Tikang
    -0.66
     supplémentaires
    -0.66
     revanche
    -0.66
    setVerticalGroup
    -0.63
    ✨:
    -0.61
     متعلقه
    -0.61
     avoient
    -0.60
    POSITIVE LOGITS
     discusses
    0.65
    })));
    0.59
     referenties
    0.57
    ])));
    0.57
     discuss
    0.56
     focus
    0.54
    };*/
    0.53
     kasarigan
    0.53
     includes
    0.52
     Topics
    0.52
    Activations Density 0.157%

    No Known Activations