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-2-9B-IT
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 13700
    Prev
    Next
    INDEX
    Explanations

    terms related to legal and organizational structures or actions

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

    "draws," "set," or "gain" followed by a preposition

    np_acts-logits-general · gemini-2.0-flash

    draw conclusions or comparisons

    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
     vôtre
    -0.34
     Danach
    -0.27
    because
    -0.27
     él
    -0.26
    at
    -0.25
    else
    -0.24
     Außerdem
    -0.24
    a
    -0.24
    if
    -0.24
    ,
    -0.24
    POSITIVE LOGITS
    ſicht
    1.13
     zwiſchen
    1.09
    ðsíða
    1.09
    <unused41>
    1.09
    iſche
    1.09
    <unused8>
    1.08
    <unused14>
    1.08
    <unused3>
    1.08
    <pad>
    1.08
    [@BOS@]
    1.08
    Activations Density 0.991%

    No Known Activations