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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & 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. 20-GEMMASCOPE-RES-16K
    6. 2924
    Prev
    Next
    INDEX
    Explanations

    specific section markers or references within texts that indicate appendices or supplementary materials

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

    appendices

    np_acts-logits-general · gemini-2.0-flash

    appendix references

    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-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71
    Prompts (Dashboard)
    36,864 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
     iconFacebook
    -0.61
    '].'
    -0.61
     disponibilités
    -0.60
    ://"
    -0.59
    ']}
    -0.58
    Cyfarwyddwr
    -0.58
    }*/
    
    -0.58
    _"+
    -0.57
    ******/
    -0.56
    _'+
    -0.56
    POSITIVE LOGITS
    TabIndex
    2.66
     مشين
    0.67
    tabIndex
    0.66
    CORBA
    0.62
    DockStyle
    0.59
    setVerticalGroup
    0.58
    ospor
    0.54
    shadowOpacity
    0.54
     tabPage
    0.53
    RetentionPolicy
    0.52
    Activations Density 0.012%

    No Known Activations