© 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. 10629
    Prev
    Next
    INDEX
    Explanations

    mentions of various vitamins and minerals in relation to nutrition deficiencies

    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
    +0.17
    1.0%
    23
    +0.14
    0.8%
    231
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    50
    +0.17
    0.09
    494
    +0.14
    0.07
    210
    +0.13
    0.05
    Negative Logits
    EMENT
    -1.85
     yourself
    -1.70
     programme
    -1.70
    iary
    -1.69
    iveness
    -1.67
     checklist
    -1.65
    iative
    -1.63
    ariate
    -1.55
    :`
    -1.53
    ivals
    -1.52
    POSITIVE LOGITS
    »¿
    3.20
    º
    3.15
    ±
    3.10
    Ļ
    3.06
    ĻĤ
    3.04
    ı
    3.03
    Ķ
    3.03
    ľ
    2.99
    ¯
    2.98
    Ŀ
    2.96
    Activations Density 0.976%

    No Known Activations