© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 11614
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    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
    45
    +0.12
    0.6%
    442
    +0.11
    0.6%
    319
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    0
    -0.12
    0.00
    1
    -0.11
    0.00
    2
    -0.10
    0.00
    Negative Logits
    ]{.
    -1.68
     thereafter
    -1.68
    nikov
    -1.47
    atra
    -1.46
    ANK
    -1.45
     filing
    -1.44
    isu
    -1.41
    ims
    -1.39
    ash
    -1.38
     chloride
    -1.35
    POSITIVE LOGITS
    taire
    1.72
    uvre
    1.39
    eland
    1.37
     champ
    1.33
     hull
    1.33
    gered
    1.32
     realised
    1.32
    ading
    1.31
     logo
    1.30
     drunk
    1.27
    Activations Density 0.000%

    No Known Activations

    This feature has no known activations.