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-2-9B
    4. Residual Stream - 16k
    5. 36-GEMMASCOPE-RES-16K
    6. 14540
    Prev
    Next
    INDEX
    Explanations

    statements about individual choices and moral dilemmas

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 36-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_36/width_16k/average_l0_120
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.36.hook_resid_post
    Hook Layer
    36
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     vraiment
    -0.50
    真是
    -0.50
     razu
    -0.43
     davvero
    -0.42
    还真是
    -0.41
     really
    -0.41
     verkligen
    -0.40
     szczególnie
    -0.39
     Truly
    -0.39
     habituales
    -0.39
    POSITIVE LOGITS
     fucking
    0.87
     goddamn
    0.79
     freakin
    0.73
     damn
    0.71
     arguably
    0.71
    fucking
    0.71
     freaking
    0.70
     dammit
    0.70
     fuckin
    0.69
     shitty
    0.68
    Activations Density 0.995%

    No Known Activations