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-2-2B
    4. Residual Stream - 16k
    5. 22-GEMMASCOPE-RES-16K
    6. 7955
    Prev
    Next
    INDEX
    Explanations

    references to opioids and their implications

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

    The token "op" and its variations

    np_acts-logits-general · gemini-2.0-flash

    words starting with op

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.hook_resid_post
    Hook Layer
    22
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    urably
    -0.52
     kanya
    -0.52
     feit
    -0.50
    ño
    -0.50
    Initializable
    -0.49
     Madison
    -0.48
     viewType
    -0.47
    Autoritní
    -0.46
     newBuilder
    -0.46
     怎样
    -0.46
    POSITIVE LOGITS
     Op
    0.77
     op
    0.73
     Winfrey
    0.67
     Opa
    0.66
     ops
    0.66
    Op
    0.65
    Composable
    0.64
     Ops
    0.62
    Ops
    0.61
     Mop
    0.61
    Activations Density 0.083%

    No Known Activations