© 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. MLP Post
    9. 2-MLP-SM
    11. 25285
    Prev
    Next
    INDEX
    Explanations

    specific numerical values or codes, likely indicating parameters or settings

    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
    156
    +0.16
    0.9%
    23
    +0.14
    0.8%
    293
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    23
    +0.16
    0.03
    56
    +0.14
    0.08
    326
    +0.11
    0.02
    Negative Logits
    úblic
    -1.63
    ansas
    -1.62
    inates
    -1.58
    $.[]{
    -1.50
    č↵č↵
    -1.50
    boro
    -1.47
    ific
    -1.37
    __,
    -1.36
    (),
    -1.36
    $/
    -1.33
    POSITIVE LOGITS
    ĻĤ
    2.39
    ±
    2.22
    ¿½
    2.08
    ij
    2.08
    Ĥ¬
    2.06
    ĥ
    2.06
    Ļª
    2.04
    ²
    1.98
    ´
    1.89
    ĨĴ
    1.86
    Activations Density 1.431%

    No Known Activations