© 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. 1-RES-SM
    11. 29690
    Prev
    Next
    INDEX
    Explanations

    references to specific protocols or configurations in data transmission systems

    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/1-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.1.hook_resid_post
    Hook Layer
    1
    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
    502
    +0.17
    1.0%
    300
    +0.13
    0.7%
    156
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    502
    +0.17
    0.03
    300
    +0.13
    0.03
    39
    +0.10
    0.03
    Negative Logits
    ĥ½
    -5.17
    Ĥ
    -4.14
    Ń
    -4.08
    ľĵ
    -4.04
    Īĺ
    -3.99
    ¿½
    -3.89
    ¼
    -3.87
    ·
    -3.82
    ĥ
    -3.77
    ©
    -3.71
    POSITIVE LOGITS
    ublic
    1.98
    arser
    1.89
    edia
    1.88
    GF
    1.83
    ND
    1.80
    NP
    1.77
    seud
    1.75
    NI
    1.72
     Films
    1.72
    NR
    1.71
    Activations Density 0.110%

    No Known Activations