© 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. 2-RES-SM
    11. 31531
    Prev
    Next
    INDEX
    Explanations

    references to causes of conditions or events

    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/2-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.2.hook_resid_post
    Hook Layer
    2
    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
    5
    +0.16
    0.9%
    283
    +0.13
    0.8%
    239
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    5
    +0.16
    0.03
    485
    +0.13
    0.02
    397
    +0.12
    0.02
    Negative Logits
    Īĺ
    -2.83
    ´
    -2.76
    ¢
    -2.60
    ²
    -2.55
    ¡
    -2.46
    ¿½
    -2.44
    µ
    -2.36
    ¤
    -2.31
    ħ
    -2.28
    £
    -2.28
    POSITIVE LOGITS
    way
    2.22
    liest
    1.98
    uric
    1.90
    ional
    1.88
    aways
    1.70
    atics
    1.68
    aret
    1.63
     pathogenesis
    1.61
    ages
    1.60
    arial
    1.59
    Activations Density 0.075%

    No Known Activations