© 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
    2. Joseph Bloom · Initial Gemma-2B-IT SAEs
    3. Gemma-2B-IT
    4. Residual Stream
    5. 12-RES-JB
    6. 6619
    Prev
    Next
    INDEX
    Explanations

    terms related to different industries and sectors, such as fitness centers, museums, agriculture, health and safety, technology, and banking

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    jbloom/Gemma-2b-IT-Residual-Stream-SAEs/gemma_2b_it_blocks.12.hook_resid_post_16384
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    HuggingFaceFW/fineweb
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.12.hook_resid_post
    Hook Layer
    12
    Architecture
    standard
    Context Size
    1,024
    Dataset
    Skylion007/openwebtext
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    50
    +0.28
    1.0%
    1577
    +0.13
    0.5%
    1108
    +0.11
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    50
    +0.28
    0.17
    1870
    +0.13
    0.09
    1856
    +0.11
    0.16
    Negative Logits
     Himo
    -0.91
     Obrador
    -0.87
    TagMode
    -0.87
     onCancelled
    -0.86
     Baillargeon
    -0.85
    RectangleBorder
    -0.84
     Unito
    -0.83
     onAnimation
    -0.83
     Bourgoin
    -0.83
     laude
    -0.82
    POSITIVE LOGITS
     McLaugh
    1.86
     impra
    1.83
     shenan
    1.78
     unspeak
    1.77
     ineffec
    1.74
     unlaw
    1.69
     intersper
    1.67
     uninten
    1.66
     Rine
    1.62
     encomp
    1.62
    Activations Density 3.682%

    No Known Activations