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. 15-GEMMASCOPE-RES-16K
    6. 5182
    Prev
    Next
    INDEX
    Explanations

    references to identity changes and the use of aliases

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.15.hook_resid_post
    Hook Layer
    15
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ACHUSET
    -0.51
     BoxFit
    -0.49
    )}</
    -0.46
    ]})
    -0.43
     gynhyrchwyd
    -0.43
    ])));
    -0.42
     relâche
    -0.42
    parentElement
    -0.41
    InStock
    -0.41
    ximity
    -0.41
    POSITIVE LOGITS
     name
    0.89
     pseudonym
    0.88
     alias
    0.83
     aliases
    0.78
    name
    0.77
     propOrder
    0.73
     surname
    0.71
     nickname
    0.69
     nome
    0.68
     naam
    0.68
    Activations Density 0.245%

    No Known Activations