Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    3. Pythia-70M-Deduped
    4. MLP Post
    5. 0-MLP-SM
    6. 27399
    Prev
    Next
    INDEX
    Explanations

    references to various types of services

    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
    5
    +0.14
    0.8%
    94
    +0.14
    0.7%
    13
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    250
    +0.14
    0.04
    80
    +0.14
    0.04
    0
    +0.11
    0.04
    Negative Logits
    ller
    -1.79
    aliana
    -1.61
    istically
    -1.59
    icable
    -1.53
    óg
    -1.51
    izing
    -1.50
    ually
    -1.49
     such
    -1.48
    ally
    -1.47
    uced
    -1.44
    POSITIVE LOGITS
    documentclass
    1.67
    srep
    1.63
     accession
    1.60
     Parliament
    1.52
     doctrine
    1.52
    mine
    1.50
    bons
    1.48
    apine
    1.47
     NATO
    1.47
    bles
    1.46
    Activations Density 0.021%

    No Known Activations