© 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. 25114
    Prev
    Next
    INDEX
    Explanations

    occurrences of data structures and type definitions in programming code

    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
    122
    +0.15
    0.8%
    411
    +0.14
    0.8%
    475
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    164
    +0.15
    0.03
    411
    +0.14
    0.02
    122
    +0.13
    0.02
    Negative Logits
    uximab
    -1.75
    ĥ½
    -1.54
    an
    -1.52
    ĺ
    -1.52
    aple
    -1.50
     \$
    -1.49
    yrus
    -1.46
    plates
    -1.41
     phosphate
    -1.37
     flank
    -1.37
    POSITIVE LOGITS
     ”
    1.51
     ‘
    1.50
    ).)
    1.45
     Lif
    1.43
     sunset
    1.41
    sson
    1.36
     “[
    1.35
     uninsured
    1.34
    stant
    1.33
     dash
    1.32
    Activations Density 0.173%

    No Known Activations