© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 16k
    5. 5-GEMMASCOPE-RES-16K
    6. 338
    Prev
    Next
    INDEX
    Explanations

    terms related to choking and suffocation

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.5.hook_resid_post
    Hook Layer
    5
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ]--;
    -0.76
    Obrázky
    -0.65
     Cari
    -0.62
    IonicModule
    -0.60
    sego
    -0.60
    ագրություններ
    -0.60
     numberOfRows
    -0.57
     väli
    -0.56
     esportes
    -0.56
    ">−
    -0.55
    POSITIVE LOGITS
     choke
    1.21
     choking
    1.14
     strangle
    1.04
     strangled
    1.03
     choked
    1.02
     strang
    0.95
     étrang
    0.94
     neck
    0.92
     suffoc
    0.91
    choke
    0.88
    Activations Density 0.409%

    No Known Activations