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-2-9B-IT
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 14910
    Prev
    Next
    INDEX
    Explanations

    commands and terms related to programming, particularly for installation and version control tasks

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

    Code snippets after shell commands

    np_acts-logits-general · gemini-2.0-flash

    shell commands and actions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    RegressionTest
    -0.69
    StoryboardSegue
    -0.59
    ValueStyle
    -0.53
    互联网档案馆
    -0.49
    ValueGeneration
    -0.48
     queſta
    -0.47
    tagHelperRunner
    -0.47
    TacToe
    -0.46
     MainAxisSize
    -0.46
    rizona
    -0.46
    POSITIVE LOGITS
     terminal
    0.37
    devServer
    0.36
     command
    0.36
     spell
    0.36
    命令
    0.34
     dry
    0.32
    cmd
    0.31
     tar
    0.31
    ResponseType
    0.31
    aliases
    0.30
    Activations Density 0.444%

    No Known Activations