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

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 29684
    Prev
    Next
    INDEX
    Explanations

    negative emotions and attitudes expressed towards various subjects

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

    character traits or manners

    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
     campeão
    -0.36
     mouseY
    -0.35
     كومونز
    -0.35
     taxas
    -0.35
     mesmas
    -0.34
    不住
    -0.33
     resourceCulture
    -0.33
    Hozzáférés
    -0.33
     TestBed
    -0.32
    enumi
    -0.31
    POSITIVE LOGITS
    GEBURTSDATUM
    0.65
     يتيمه
    0.59
    rungsseite
    0.55
    TargetApi
    0.46
    meuble
    0.46
     ioutil
    0.45
    umenical
    0.45
     barbati
    0.45
     uſed
    0.45
     raiſ
    0.45
    Activations Density 0.058%

    No Known Activations