© 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. 24-GEMMASCOPE-RES-16K
    6. 15914
    Prev
    Next
    INDEX
    Explanations

    terminologies and phrases related to surgical procedures and anatomical structures

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

    Removal or detaching something

    np_acts-logits-general · gemini-2.0-flash

    removed or being removed

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 24-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_24/width_16k/average_l0_73
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.24.hook_resid_post
    Hook Layer
    24
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     gynnwys
    -0.61
    WriteBarrier
    -0.55
     kors
    -0.53
    MergeFrom
    -0.50
    arynx
    -0.50
    hnung
    -0.49
     Encode
    -0.47
    ConverterFactory
    -0.47
    criminate
    -0.47
    UrlEncoded
    -0.46
    POSITIVE LOGITS
     removal
    1.94
     removed
    1.91
     remove
    1.84
     Removal
    1.73
     removing
    1.73
     Remove
    1.61
     removes
    1.59
     REMOV
    1.58
     Removing
    1.57
    removal
    1.57
    Activations Density 0.471%

    No Known Activations