© 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 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. 13066
    Prev
    Next
    INDEX
    Explanations

    references to magnetic fields and their properties

    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/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
    443
    +0.15
    0.9%
    346
    +0.15
    0.9%
    129
    +0.13
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    443
    +0.15
    0.01
    501
    +0.15
    0.01
    346
    +0.13
    0.01
    Negative Logits
    ership
    -2.02
    sible
    -1.76
    ioned
    -1.76
    nai
    -1.73
    umerable
    -1.72
    angered
    -1.57
    ophageal
    -1.57
    onym
    -1.56
    eful
    -1.51
    ément
    -1.47
    POSITIVE LOGITS
     resonance
    2.30
     dipole
    1.89
    agnet
    1.71
    Field
    1.64
    field
    1.62
    rif
    1.61
    romagnetic
    1.55
    arm
    1.54
     braking
    1.53
     field
    1.52
    Activations Density 0.034%

    No Known Activations