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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 65k
    5. 13-GEMMASCOPE-RES-65K
    6. 56407
    Prev
    Next
    INDEX
    Explanations

    terms related to "forward" and "backward" motions or statements, particularly in contexts involving projections or predictions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 13-gemmascope-res-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_13/width_65k/average_l0_75
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    Data Type
    float32
    Hook Name
    blocks.13.hook_resid_post
    Hook Layer
    13
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    didSet
    -0.90
    ibrill
    -0.85
    JSONException
    -0.85
    onAttach
    -0.83
    AnchorTagHelper
    -0.82
    ądź
    -0.81
    getColumnIndex
    -0.81
     समीक्षाओं
    -0.80
    MeasureSpec
    -0.79
     violenza
    -0.79
    POSITIVE LOGITS
     Forward
    1.56
     FORWARD
    1.54
    Forward
    1.52
     forwards
    1.45
     forward
    1.38
    forward
    1.38
    FORWARD
    1.34
     fow
    1.20
     forwarding
    1.18
     fwd
    1.01
    Activations Density 0.016%

    No Known Activations