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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 3-GEMMASCOPE-RES-16K
    6. 12716
    Prev
    Next
    INDEX
    Explanations

    references to adversarial dynamics or competitors in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 3-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_3/width_16k/average_l0_59
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.3.hook_resid_post
    Hook Layer
    3
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Ku
    -0.55
    вести
    -0.54
    ide
    -0.54
     be
    -0.53
     Bara
    -0.53
     flux
    -0.52
    </td>
    -0.52
    Bara
    -0.50
    saw
    -0.50
    bara
    -0.50
    POSITIVE LOGITS
     opponent
    2.36
     opponents
    2.13
    opponent
    1.85
    Opponent
    1.71
     Gegner
    1.49
     Opp
    1.48
    onents
    1.48
    onent
    1.36
    PONENT
    1.29
     adversary
    1.25
    Activations Density 0.102%

    No Known Activations