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. 18-GEMMASCOPE-RES-16K
    6. 1014
    Prev
    Next
    INDEX
    Explanations

    commands and functions related to file manipulation and scripting in a programming context

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

    Bash script commands/syntax

    np_acts-logits-general · gemini-2.0-flash

    shell script `if` statements

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.18.hook_resid_post
    Hook Layer
    18
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     mxArray
    -0.53
    sprit
    -0.51
    Хьажоргаш
    -0.51
    elsif
    -0.48
    endif
    -0.48
    endregion
    -0.48
     OnInit
    -0.48
    astify
    -0.47
    ngOnInit
    -0.47
    fromCharCode
    -0.47
    POSITIVE LOGITS
     متعلقه
    0.80
     echo
    0.59
    
    0.55
     allAfrica
    0.54
     gawas
    0.53
     \\\
    0.53
     asmen
    0.53
    AntiForgeryToken
    0.52
    RegistryLite
    0.52
     ProtoMessage
    0.52
    Activations Density 0.737%

    No Known Activations