© 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. 24885
    Prev
    Next
    INDEX
    Explanations

    code structure and iteration commands in programming languages

    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
    176
    +0.15
    0.9%
    70
    +0.13
    0.7%
    386
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    70
    +0.15
    0.04
    435
    +0.13
    0.03
    170
    +0.12
    0.03
    Negative Logits
    à±ģ
    -1.54
    ersion
    -1.48
    pei
    -1.46
    izi
    -1.46
    omycin
    -1.45
    rika
    -1.45
    blic
    -1.43
    ikan
    -1.43
    ideo
    -1.43
    oltzmann
    -1.41
    POSITIVE LOGITS
     starters
    1.46
    frontal
    1.43
     possessions
    1.42
    elves
    1.39
     siblings
    1.39
     """
    1.36
     secrets
    1.36
    """
    1.36
     sink
    1.34
    (#
    1.33
    Activations Density 0.462%

    No Known Activations