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-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 10749
    Prev
    Next
    INDEX
    Explanations

    terms related to discrimination in employment practices

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

    protected characteristics

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    MLLoader
    -0.70
    parsedMessage
    -0.66
     utafitiHapana
    -0.63
    المكان
    -0.61
     surla
    -0.61
    msgTypes
    -0.59
    帖最后由
    -0.59
     ErrIntOverflow
    -0.59
    /**
    -0.58
    تقاوى
    -0.56
    POSITIVE LOGITS
     condena
    0.32
     Herkunft
    0.32
     económico
    0.32
     bailarina
    0.32
     TestBed
    0.31
    mentación
    0.31
     incluyen
    0.31
    denza
    0.31
     or
    0.31
     usia
    0.30
    Activations Density 0.018%

    No Known Activations