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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Initial Gemma-2B-IT SAEs
    5. Gemma-2B-IT
    7. Residual Stream
    9. 12-RES-JB
    11. 5465
    Prev
    Next
    INDEX
    Explanations

    references to characters and elements related to Batman comics

    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
    662
    +0.19
    0.9%
    1527
    +0.15
    0.7%
    1034
    +0.14
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    662
    +0.19
    0.05
    227
    +0.15
    0.04
    1177
    +0.14
    0.02
    Negative Logits
     disambiguazione
    -0.59
    marginHorizontal
    -0.53
    calaure
    -0.52
     Holbrook
    -0.50
    adalajara
    -0.50
    beforeSend
    -0.49
    pses
    -0.49
    Rudi
    -0.48
     للاسماء
    -0.48
    Concord
    -0.46
    POSITIVE LOGITS
     Knight
    1.05
    Knight
    0.99
     knight
    0.98
     Wayne
    0.98
     knights
    0.95
    knight
    0.92
    Wayne
    0.91
     Knights
    0.90
     clay
    0.89
    Knights
    0.88
    Activations Density 0.280%

    No Known Activations