© 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. 25719
    Prev
    Next
    INDEX
    Explanations

    structured data format elements, particularly key-value pairs

    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
    238
    +0.15
    0.8%
    223
    +0.12
    0.7%
    12
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    238
    +0.15
    0.03
    17
    +0.12
    0.01
    102
    +0.12
    0.03
    Negative Logits
    ĻĤ
    -1.73
    «
    -1.69
    ctor
    -1.69
    Īĺ
    -1.63
    reflect
    -1.55
    Ļ
    -1.50
    structure
    -1.48
    aqu
    -1.48
    close
    -1.48
    rens
    -1.43
    POSITIVE LOGITS
    terday
    1.62
    atonin
    1.61
    lights
    1.57
    aways
    1.53
     cigarettes
    1.51
     headaches
    1.46
    thood
    1.46
     facts
    1.43
    enance
    1.42
    lle
    1.42
    Activations Density 0.375%

    No Known Activations