© 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 EvalsExportsinterp-engineNEWAPI 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. 16122
    Prev
    Next
    INDEX
    Explanations

    mentions of groupings of people or individuals with specific roles within organizations

    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
    227
    +0.14
    0.4%
    1919
    +0.12
    0.4%
    394
    +0.12
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1919
    +0.14
    0.08
    862
    +0.12
    0.04
    227
    +0.12
    0.07
    Negative Logits
     silikon
    -1.04
     mikrofon
    -0.97
     poliester
    -0.96
     antik
    -0.95
     torba
    -0.92
     balon
    -0.91
     brutto
    -0.89
     makro
    -0.88
     persil
    -0.88
     alkoh
    -0.85
    POSITIVE LOGITS
     unwarran
    0.87
     apprehen
    0.75
     themselves
    0.75
     tolerably
    0.75
     gaily
    0.73
     are
    0.72
     disagre
    0.72
     shewn
    0.70
     withal
    0.70
     nobly
    0.70
    Activations Density 0.395%

    No Known Activations