© 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. 2-RES-SM
    11. 9777
    Prev
    Next
    INDEX
    Explanations

    terms and phrases related to viral infections and medical conditions

    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
    156
    +0.24
    1.4%
    74
    +0.11
    0.6%
    345
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    446
    +0.24
    0.03
    186
    +0.11
    0.03
    254
    +0.10
    0.04
    Negative Logits
    roe
    -1.82
    ori
    -1.65
    astern
    -1.61
    oscopic
    -1.60
    else
    -1.59
     mismo
    -1.57
    atsu
    -1.53
    aring
    -1.53
     consider
    -1.48
    owing
    -1.47
    POSITIVE LOGITS
    ¢
    3.87
    £
    3.80
    µ
    3.65
    ĸ
    3.59
    ¬
    3.55
    §
    3.46
    Ń
    3.40
    ¦
    3.38
    ¥
    3.36
    ¾
    3.34
    Activations Density 0.366%

    No Known Activations