Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & 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. 21-GEMMASCOPE-RES-16K
    6. 1026
    Prev
    Next
    INDEX
    Explanations

    block start or end markers in text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 21-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_21/width_16k/average_l0_70
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.21.hook_resid_post
    Hook Layer
    21
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    DIPSETTING
    -0.58
    posób
    -0.53
    ütfen
    -0.53
    WriteBarrier
    -0.51
    ärna
    -0.50
    świę
    -0.50
    extAlignment
    -0.50
    ึ้น
    -0.50
    httphttps
    -0.50
     keinem
    -0.49
    POSITIVE LOGITS
    According
    0.87
    Speaking
    0.82
     Speaking
    0.81
     According
    0.80
    Meanwhile
    0.79
     Meanwhile
    0.73
    Commenting
    0.70
     Selon
    0.67
     according
    0.66
    Rea
    0.66
    Activations Density 0.026%

    No Known Activations