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. 13003
    Prev
    Next
    INDEX
    Explanations

    connections between species, particularly regarding their traits and health implications

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

    "human" or "humans"

    np_acts-logits-general · gemini-2.0-flash

    human and man

    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
     везе
    -0.52
     }}"></
    -0.48
    ')")
    -0.48
     isoto
    -0.48
    IsContent
    -0.47
     martyrdom
    -0.46
     yonder
    -0.46
    rency
    -0.46
    ISTRO
    -0.45
    estero
    -0.45
    POSITIVE LOGITS
     human
    3.53
     humans
    3.12
    human
    3.09
    Human
    2.86
     Human
    2.77
     HUMAN
    2.72
     humain
    2.63
     humano
    2.61
     Humans
    2.60
     humanos
    2.59
    Activations Density 0.529%

    No Known Activations