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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 16k
    5. 9-GEMMASCOPE-RES-16K
    6. 3287
    Prev
    Next
    INDEX
    Explanations

    structural and syntactical elements of code, particularly relating to list slicing and indexing in programming languages

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

    Slicing or extracting substrings from a string

    np_acts-logits-general · gemini-2.0-flash

    slicing and substrings

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_9/width_16k/average_l0_88
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.9.hook_resid_post
    Hook Layer
    9
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    +:+
    -0.57
    tagHelperRunner
    -0.49
     invokingState
    -0.49
     nonUne
    -0.48
    KommentareTeilen
    -0.46
    writerow
    -0.46
    Erreferentziak
    -0.45
    DoubleQuotes
    -0.45
    uxxxx
    -0.45
     Ril
    -0.44
    POSITIVE LOGITS
    substring
    0.92
    slice
    0.80
    substr
    0.71
    Substring
    0.69
    Slice
    0.56
     substring
    0.51
     slice
    0.51
     Slice
    0.49
    DockStyle
    0.47
     截
    0.47
    Activations Density 0.239%

    No Known Activations