© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B
    7. Residual Stream - 16k
    9. 31-GEMMASCOPE-RES-16K
    11. 6243
    Prev
    Next
    INDEX
    Explanations

    phrases that indicate conjecture, belief, or propositions regarding outcomes and observations

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

    causal or conditional clauses introduced by "such that".

    oai_token-act-pair · claude-4-5-haikuTriggered by @emiglarou
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 31-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_31/width_16k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.31.hook_resid_post
    Hook Layer
    31
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    GOTREF
    -0.74
    addGap
    -0.69
    MLLoader
    -0.62
    __*/
    -0.62
     оригіналу
    -0.60
    ItemBackground
    -0.59
    TargetException
    -0.59
    ContentAsync
    -0.58
    addPreferredGap
    -0.57
    QMetaType
    -0.57
    POSITIVE LOGITS
     that
    2.53
     bahwa
    1.90
     że
    1.52
     rằng
    1.51
    ว่า
    1.47
     bahawa
    1.46
     kwamba
    1.41
     että
    1.34
     ότι
    1.32
     dass
    1.30
    Activations Density 2.752%

    No Known Activations