© 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. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. Residuals
    9. 2-RES-SM
    11. 32080
    Prev
    Next
    INDEX
    Explanations

    warnings or alerts related to potential dangers or risks

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    ctigges/pythia-70m-deduped__res-sm_processed/2-res-sm
    Prompts (Dashboard)
    32,768 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.2.hook_resid_post
    Hook Layer
    2
    Architecture
    standard
    Context Size
    128
    Dataset
    EleutherAI/the_pile_deduplicated
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    369
    +0.17
    1.0%
    497
    +0.13
    0.7%
    91
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    329
    +0.17
    0.03
    411
    +0.13
    0.02
    167
    +0.12
    0.02
    Negative Logits
    oden
    -1.54
     resulted
    -1.47
     concern
    -1.45
    ounder
    -1.45
    lessness
    -1.44
    oding
    -1.44
    oxin
    -1.43
    esting
    -1.42
    osion
    -1.37
    cribing
    -1.35
    POSITIVE LOGITS
    Ĥ¬
    1.79
    Īĺ
    1.77
    ·
    1.71
    IJ
    1.62
    Ĵ
    1.62
    ¡
    1.60
    ĥ½
    1.60
    Ļª
    1.57
    ŀ
    1.57
     android
    1.51
    Activations Density 0.195%

    No Known Activations