© 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. 25497
    Prev
    Next
    INDEX
    Explanations

    references to dictionaries or dictionary-related concepts

    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
    +0.24
    1.4%
    380
    +0.13
    0.7%
    110
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    379
    +0.24
    0.02
    392
    +0.13
    0.02
    380
    +0.11
    0.01
    Negative Logits
    beit
    -1.74
     reasonable
    -1.71
     sudden
    -1.67
     particular
    -1.58
     substantial
    -1.57
    opropyl
    -1.56
     September
    -1.53
     certain
    -1.49
     due
    -1.47
     significant
    -1.47
    POSITIVE LOGITS
    arters
    1.65
    cias
    1.62
    holder
    1.61
    estry
    1.60
    AtIndex
    1.56
    buk
    1.55
    etable
    1.51
    iterator
    1.49
    lists
    1.47
    ctl
    1.47
    Activations Density 0.113%

    No Known Activations