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 - 65k
    5. 20-GEMMASCOPE-RES-65K
    6. 749
    Prev
    Next
    INDEX
    Explanations

    syntactical elements and structure in programming code

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

    programming-related syntax elements, particularly those related to function calls and control structures

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

    have not

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_65k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    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
     immédi
    -0.60
    nameof
    -0.54
    thâu
    -0.53
     immediate
    -0.51
     ErrInvalid
    -0.49
     inkább
    -0.49
    mybatisplus
    -0.49
     Immediate
    -0.49
    řed
    -0.47
     vero
    -0.46
    POSITIVE LOGITS
     endfor
    0.93
    endfor
    0.83
     endforeach
    0.77
    hoeddwyd
    0.72
    AddTagHelper
    0.71
    帖最后由
    0.69
    Sucesor
    0.67
    endforeach
    0.66
     AssemblyCulture
    0.64
    continue
    0.57
    Activations Density 0.019%

    No Known Activations