© 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. 0-MLP-SM
    11. 28850
    Prev
    Next
    INDEX
    Explanations

    terms related to softness

    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
    172
    +0.12
    0.7%
    59
    +0.12
    0.7%
    99
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    57
    +0.12
    0.02
    452
    +0.12
    0.03
    172
    +0.11
    0.02
    Negative Logits
    iculously
    -1.67
    ĥ
    -1.59
     hers
    -1.57
    rost
    -1.55
     most
    -1.46
    ropolis
    -1.44
    ĨĴ
    -1.43
    rog
    -1.42
    Į
    -1.42
    umatic
    -1.42
    POSITIVE LOGITS
    idelines
    1.86
    stown
    1.84
    ener
    1.82
    est
    1.68
    ening
    1.63
    ened
    1.54
    itud
    1.49
    ball
    1.49
    hearted
    1.48
    ified
    1.46
    Activations Density 0.017%

    No Known Activations