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-2-9B
    4. Residual Stream - 16k
    5. 21-GEMMASCOPE-RES-16K
    6. 14380
    Prev
    Next
    INDEX
    Explanations

    themes related to emotional experiences and social interactions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129
    Prompts (Dashboard)
    24,576 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
    enumi
    -0.64
     Савезне
    -0.46
     Zeller
    -0.45
    üyada
    -0.43
    帖最后由
    -0.41
     potrafi
    -0.40
    seits
    -0.40
    ValueStyle
    -0.40
    routeProvider
    -0.40
     olmayan
    -0.40
    POSITIVE LOGITS
    +#+
    0.55
     judgment
    0.50
     cart
    0.47
     judgement
    0.46
    ientras
    0.45
    ioutil
    0.44
     phon
    0.44
     conclusion
    0.44
     emotion
    0.44
     dere
    0.43
    Activations Density 0.948%

    No Known Activations