© 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. 25497
    Prev
    Next
    INDEX
    Explanations

    the phrase "According to" indicating references or citations in texts

    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%
    210
    +0.12
    0.7%
    502
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    210
    +0.16
    0.03
    367
    +0.12
    0.02
    349
    +0.12
    0.02
    Negative Logits
     committed
    -1.58
     devoted
    -1.49
     dab
    -1.47
    TY
    -1.45
    igating
    -1.43
     charges
    -1.31
     darling
    -1.31
     engaged
    -1.29
     studied
    -1.29
    joy
    -1.28
    POSITIVE LOGITS
    emente
    1.60
    chrotron
    1.59
    placement
    1.53
    endor
    1.52
    below
    1.51
    enza
    1.51
    agraph
    1.47
    tons
    1.47
    ¾
    1.44
    orden
    1.43
    Activations Density 0.018%

    No Known Activations