© 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. 25545
    Prev
    Next
    INDEX
    Explanations

    references to various tools and resources used for support and optimization

    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
    156
    +0.26
    1.5%
    184
    +0.13
    0.7%
    173
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    184
    +0.26
    0.03
    295
    +0.13
    0.02
    95
    +0.12
    0.02
    Negative Logits
    §
    -2.07
    »¿
    -2.02
    áĢ¯
    -2.00
    ³
    -1.91
    ı
    -1.88
    ATES
    -1.84
    ĥ½
    -1.82
    Ĭ
    -1.81
    ©
    -1.81
    ¨
    -1.77
    POSITIVE LOGITS
    mith
    2.42
    olver
    2.05
    etting
    1.98
    pace
    1.92
    urgical
    1.82
    point
    1.81
    heet
    1.76
    helf
    1.71
    kin
    1.71
    hell
    1.68
    Activations Density 0.105%

    No Known Activations