© 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-2-9B
    4. Residual Stream - 16k
    5. 24-GEMMASCOPE-RES-16K
    6. 5816
    Prev
    Next
    INDEX
    Explanations

    terms and phrases related to legal case assignments and errors

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

    "ASSIGNMENT OF ERROR NUMBER" phrases

    np_acts-logits-general · gemini-2.0-flash

    assignments of error or supplementary movies

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 24-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_24/width_16k/average_l0_114
    Prompts (Dashboard)
    24,576 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
     kár
    -0.33
    差し
    -0.32
     nada
    -0.32
    Schweizer
    -0.32
     ['./
    -0.32
     nothing
    -0.32
     otherwise
    -0.30
    嬸
    -0.30
     улица
    -0.30
    none
    -0.29
    POSITIVE LOGITS
     fourth
    0.68
     fifth
    0.65
     third
    0.65
     seventh
    0.64
     sixth
    0.62
    第
    0.60
     betweenstory
    0.59
     thứ
    0.58
    AnimationsModule
    0.58
    jsxFileName
    0.57
    Activations Density 1.634%

    No Known Activations