© 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. 4-RES-SM
    11. 10113
    Prev
    Next
    INDEX
    Explanations

    references to software licenses and their versions

    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/4-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.4.hook_resid_post
    Hook Layer
    4
    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
    126
    +0.13
    0.8%
    397
    +0.13
    0.8%
    450
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    126
    +0.13
    -0.00
    255
    +0.13
    -0.01
    78
    +0.12
    -0.01
    Negative Logits
    ugin
    -1.49
     ordinary
    -1.49
    ppler
    -1.42
    ropolis
    -1.40
    roit
    -1.33
    venir
    -1.32
    ypse
    -1.31
    sect
    -1.29
     proportions
    -1.27
    valence
    -1.27
    POSITIVE LOGITS
    ŀ
    3.69
    Ļ
    3.55
    Ĩ
    3.52
    ¾
    3.41
    ģ
    3.35
    ↵
    3.33
    3.33
     ↵
    3.33
    ↵    ↵
    3.33
    č↵
    3.33
    Activations Density 0.116%

    No Known Activations