© 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. 25450
    Prev
    Next
    INDEX
    Explanations

    terms related to medical and biological conditions, specifically focused on anemia and its effects

    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
    126
    +0.11
    0.6%
    177
    +0.11
    0.6%
    503
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    48
    +0.11
    0.03
    137
    +0.11
    0.02
    503
    +0.10
    0.01
    Negative Logits
     realized
    -1.69
     recognized
    -1.65
     such
    -1.63
    aming
    -1.63
     dense
    -1.62
     classified
    -1.61
     accepted
    -1.54
     regarded
    -1.53
     distant
    -1.52
     open
    -1.51
    POSITIVE LOGITS
    ität
    1.91
    illian
    1.81
    arcin
    1.72
    urity
    1.66
    lite
    1.66
    keit
    1.63
    Inflater
    1.63
    elta
    1.62
    iliary
    1.62
    rette
    1.59
    Activations Density 0.167%

    No Known Activations