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 - 131k
    5. 32-GEMMASCOPE-RES-131K
    6. 129928
    Prev
    Next
    INDEX
    Explanations

    references to dishonesty and critiques of societal narratives regarding race

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

    delusional, whining, or insecure people

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-9b-pt-res/layer_32/width_131k/average_l0_88
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.32.hook_resid_post
    Hook Layer
    32
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ValueStyle
    -0.55
    UserScript
    -0.44
    ۜ
    -0.43
    Kanpo
    -0.42
    richTextPanel
    -0.40
    .*")]
    -0.40
    givings
    -0.39
     början
    -0.39
    toHaveBeenCalled
    -0.38
    دانشنامهٔ
    -0.38
    POSITIVE LOGITS
     delusional
    0.80
     pathetic
    0.74
     inferiority
    0.69
     whining
    0.68
     jealous
    0.66
     immature
    0.66
     insecure
    0.66
     deluded
    0.66
     troll
    0.64
     childish
    0.64
    Activations Density 0.088%

    No Known Activations