© 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. 28664
    Prev
    Next
    INDEX
    Explanations

    references to legal compliance and licensing

    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
    478
    +0.23
    1.4%
    444
    +0.10
    0.6%
    184
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    473
    +0.23
    0.02
    184
    +0.10
    0.02
    116
    +0.10
    0.02
    Negative Logits
    §
    -1.69
    Ł
    -1.66
    Bigr
    -1.56
    Ń
    -1.49
    ¬
    -1.47
    µ
    -1.46
     crore
    -1.45
    Ģ
    -1.44
    ¤
    -1.43
    ²
    -1.42
    POSITIVE LOGITS
     Redistributions
    1.79
    iamo
    1.75
    oslav
    1.71
    iah
    1.68
    iar
    1.54
    iam
    1.46
    ÑģÑı
    1.43
     detailed
    1.42
     .=
    1.41
    ÑģÑĮ
    1.41
    Activations Density 0.058%

    No Known Activations