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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 31-GEMMASCOPE-RES-16K
    6. 3610
    Prev
    Next
    INDEX
    Explanations

    notations or characters indicative of a formatting style or typographic symbols

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

    hyphenated words or compound words with hyphens connecting word parts.

    oai_token-act-pair · claude-4-5-haikuTriggered by @emiglarou
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 31-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_31/width_16k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.31.hook_resid_post
    Hook Layer
    31
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     snippetHide
    -0.67
     kasarigan
    -0.66
    SharedDtor
    -0.63
    MessageOf
    -0.56
     ब्रेकडाउन
    -0.55
    \{\\
    -0.54
    RegistryLite
    -0.54
    WriteBarrier
    -0.53
    utilisons
    -0.53
     themſelves
    -0.53
    POSITIVE LOGITS
    ‐
    0.65
    ־
    0.45
    ‑
    0.43
    brigens
    0.41
    Ͳ
    0.40
    鄄
    0.40
    -​
    0.40
     ‐
    0.40
    ﹣
    0.36
    honneur
    0.35
    Activations Density 0.539%

    No Known Activations