© 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. MLP Post
    9. 0-MLP-SM
    11. 30331
    Prev
    Next
    INDEX
    Explanations

    instances of a specific coding pattern or value definition

    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
    369
    +0.13
    0.8%
    134
    +0.11
    0.7%
    16
    +0.09
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    134
    +0.13
    0.01
    358
    +0.11
    0.01
    77
    +0.09
    0.01
    Negative Logits
     sense
    -1.76
     ?"
    -1.69
     broken
    -1.56
    '?"
    -1.51
     closest
    -1.51
     small
    -1.47
     degraded
    -1.46
     ,"
    -1.45
     ];
    -1.42
     tolerated
    -1.41
    POSITIVE LOGITS
    aho
    1.80
    ruptcy
    1.75
    ãĥ¼ãĤ¸
    1.69
    bies
    1.63
    ÅĽci
    1.62
    itations
    1.61
    ery
    1.53
    lege
    1.51
    ister
    1.51
    #:
    1.49
    Activations Density 0.012%

    No Known Activations