© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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. 27845
    Prev
    Next
    INDEX
    Explanations

    mentions of the word "principle" and its variations

    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
    111
    +0.13
    0.7%
    410
    +0.11
    0.6%
    449
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    111
    +0.13
    0.02
    510
    +0.11
    0.01
    320
    +0.11
    0.01
    Negative Logits
    ")]
    -1.70
    rian
    -1.47
    tt
    -1.46
    olerance
    -1.45
     fine
    -1.44
    ctica
    -1.43
    athing
    -1.42
    uther
    -1.41
    cology
    -1.38
     minute
    -1.38
    POSITIVE LOGITS
    iples
    2.06
    etable
    1.82
    ipal
    1.82
    enos
    1.75
    ulators
    1.73
    bugs
    1.73
    iple
    1.69
    lemen
    1.63
    imately
    1.60
    orest
    1.59
    Activations Density 0.032%

    No Known Activations