© 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. 24765
    Prev
    Next
    INDEX
    Explanations

    references to cancer and its types

    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
    +0.16
    1.0%
    397
    +0.15
    0.9%
    321
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    111
    +0.16
    0.03
    271
    +0.15
    0.01
    156
    +0.12
    0.02
    Negative Logits
    Ŀ
    -1.84
    ¯
    -1.82
     maternal
    -1.80
    unreadable
    -1.79
     hers
    -1.78
     unequal
    -1.58
    ÑĤÑĮ
    -1.57
     arbitrary
    -1.52
    rfloor
    -1.51
    maha
    -1.49
    POSITIVE LOGITS
     risks
    1.71
    scape
    1.70
    icidal
    1.68
    ous
    1.59
    icide
    1.58
     centers
    1.56
    ogenicity
    1.55
    mong
    1.49
     stem
    1.49
     prevention
    1.47
    Activations Density 0.180%

    No Known Activations