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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 33-GEMMASCOPE-RES-16K
    6. 7842
    Prev
    Next
    INDEX
    Explanations

    symbols and notation used for mathematical expressions or operations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 33-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_33/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.33.hook_resid_post
    Hook Layer
    33
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    awtextra
    -0.52
    writeFieldEnd
    -0.51
    RTEE
    -0.46
    StructEnd
    -0.45
    providedIn
    -0.45
    IUrlHelper
    -0.44
    AISSEE
    -0.42
     rękaw
    -0.41
     noDo
    -0.40
     TÉCN
    -0.40
    POSITIVE LOGITS
    ++++++++++++++++
    0.89
    /−
    0.79
    ++++++++
    0.77
    ++++
    0.77
    _+
    0.74
    &+
    0.74
    +++
    0.73
    +++++
    0.73
    *+
    0.61
    /+
    0.58
    Activations Density 0.510%

    No Known Activations