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. 20-GEMMASCOPE-RES-16K
    6. 13829
    Prev
    Next
    INDEX
    Explanations

    expressions of mental instability and emotions related to sanity

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

    "crazy" or "insane"

    np_acts-logits-general · gemini-2.0-flash

    crazy and insane states

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    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
     phụ
    -0.48
    Disliked
    -0.46
     clasped
    -0.46
    TagHelper
    -0.43
    Sil
    -0.43
    MetaType
    -0.43
     Genti
    -0.43
     alder
    -0.43
    getInputStream
    -0.42
     bottes
    -0.42
    POSITIVE LOGITS
     crazy
    1.84
     insane
    1.74
     madness
    1.73
     insanity
    1.71
    crazy
    1.71
     mad
    1.58
     Crazy
    1.56
    Crazy
    1.56
    insane
    1.53
     lunatic
    1.52
    Activations Density 0.206%

    No Known Activations