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-IT
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 1055
    Prev
    Next
    INDEX
    Explanations

    punctuation marks and their distribution in the text

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

    Enumerated steps in a list

    np_acts-logits-general · gemini-2.0-flash

    lists of steps or numbered items

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    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
    SequentialGroup
    -0.53
     continúas
    -0.52
     éc
    -0.52
    httphttps
    -0.52
    IntoConstraints
    -0.50
    feier
    -0.47
    Viited
    -0.47
    Scénario
    -0.47
     doInBackground
    -0.47
    ագրություններ
    -0.46
    POSITIVE LOGITS
    <bos>
    0.58
     the
    0.58
    └──
    0.56
     ͡°)
    0.49
    The
    0.48
    item
    0.46
    ├──
    0.45
    __(/*!
    0.44
     "..\..\..\
    0.43
    |.
    0.42
    Activations Density 0.183%

    No Known Activations