© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B
    7. Residual Stream - 16k
    9. 20-GEMMASCOPE-RES-16K
    11. 11459
    Prev
    Next
    INDEX
    Explanations

    quantifiable attributes and procedural descriptions related to technical implementations

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

    sentences that conclude or summarize technical explanations, often appearing at the end of paragraphs.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    sentence boundaries and discourse connectives that link or contrast clauses (e.g., marking ends of sentences and transitions in explanatory text).

    oai_token-act-pair · gpt-5Triggered by @jyhe0408

    periods at the end of sentences in technical or scientific text.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_16k/average_l0_68
    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
     fubject
    -0.57
     laſſen
    -0.56
     müſſen
    -0.53
     مرئيه
    -0.51
     témoig
    -0.51
     fhort
    -0.50
    onOptions
    -0.50
    LLocation
    -0.49
     oprot
    -0.49
     abſ
    -0.49
    POSITIVE LOGITS
     either
    0.39
    utilisons
    0.37
    เท่า
    0.36
     For
    0.35
     Firstly
    0.34
    Firstly
    0.34
    either
    0.33
    First
    0.32
    For
    0.31
     First
    0.31
    Activations Density 0.315%

    No Known Activations