© 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-2-9B
    4. Residual Stream - 16k
    5. 33-GEMMASCOPE-RES-16K
    6. 655
    Prev
    Next
    INDEX
    Explanations

    specialized coding constructs and structures in programming or data formats

    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
     kaarangay
    -0.88
    Autoritní
    -0.75
    ſſung
    -0.75
     AssemblyCulture
    -0.74
     dieſe
    -0.73
    ſicht
    -0.73
    iſche
    -0.71
    ፩
    -0.71
    باردا
    -0.71
    <unused16>
    -0.70
    POSITIVE LOGITS
     AssertionError
    0.35
    ://
    0.32
    <strong>
    0.31
     اینکه
    0.30
    �
    0.30
     ブルゾン
    0.29
     weiterer
    0.28
    very
    0.28
     vienas
    0.28
     procura
    0.28
    Activations Density 2.765%

    No Known Activations