© 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
    2. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    3. Pythia-70M-Deduped
    4. MLP Post
    5. 0-MLP-SM
    6. 30523
    Prev
    Next
    INDEX
    Explanations

    references to specific types of food or culinary items, particularly cheese

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    ctigges/pythia-70m-deduped__mlp-sm_processed/0-mlp-sm
    Prompts (Dashboard)
    32,768 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.0.hook_mlp_out
    Hook Layer
    0
    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
    376
    +0.16
    0.9%
    276
    +0.13
    0.7%
    241
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    276
    +0.16
    0.01
    9
    +0.13
    0.01
    453
    +0.11
    0.01
    Negative Logits
    rians
    -1.60
     themselves
    -1.50
    andum
    -1.48
    rapeutics
    -1.44
    ANS
    -1.44
    yx
    -1.44
    ?).
    -1.43
    tis
    -1.41
    OVE
    -1.38
    aned
    -1.38
    POSITIVE LOGITS
    ening
    2.23
    ÅĽci
    1.97
    istic
    1.80
    estock
    1.79
    itical
    1.78
    erie
    1.73
    velt
    1.71
    enstein
    1.70
    vre
    1.69
    ist
    1.67
    Activations Density 0.008%

    No Known Activations