© 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. 28838
    Prev
    Next
    INDEX
    Explanations

    the abbreviation "uf" and its variations, indicating a focus on specific file types or systems in software contexts

    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
    43
    +0.12
    0.7%
    283
    +0.12
    0.7%
    356
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    43
    +0.12
    0.01
    283
    +0.12
    0.01
    462
    +0.11
    0.01
    Negative Logits
     mean
    -1.78
     therefore
    -1.61
     ?"
    -1.57
    hip
    -1.54
     matter
    -1.53
     refer
    -1.52
     slightest
    -1.49
    izumab
    -1.49
     alike
    -1.48
     depend
    -1.46
    POSITIVE LOGITS
    enstein
    2.26
    thouse
    1.87
    ière
    1.74
    stadt
    1.74
    court
    1.61
    hetto
    1.60
    nell
    1.55
    ér
    1.51
    ubuntu
    1.50
    illion
    1.49
    Activations Density 0.015%

    No Known Activations