© 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. 2-MLP-SM
    6. 23714
    Prev
    Next
    INDEX
    Explanations

    elements linked to programming and data manipulation

    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/2-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.2.hook_mlp_out
    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
    97
    +0.13
    0.7%
    20
    +0.12
    0.7%
    99
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    279
    +0.13
    0.06
    23
    +0.12
    0.02
    163
    +0.10
    -0.00
    Negative Logits
    ento
    -1.90
    âĢIJ
    -1.62
     afforded
    -1.58
    edia
    -1.56
     moth
    -1.47
    ienn
    -1.46
    leen
    -1.45
     hers
    -1.42
     themselves
    -1.42
    cast
    -1.41
    POSITIVE LOGITS
    rVert
    1.60
    });
    1.60
    ory
    1.50
    $}
    1.45
     Solution
    1.41
    ']);
    1.38
     Conclusions
    1.38
    ]);
    1.35
     Conservative
    1.35
     outstanding
    1.35
    Activations Density 1.831%

    No Known Activations