© 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 EvalsExportsinterp-engineNEWAPI 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. 23386
    Prev
    Next
    INDEX
    Explanations

    references to scientific measurements and methodologies

    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
    365
    +0.12
    0.7%
    344
    +0.12
    0.7%
    343
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    125
    +0.12
    0.04
    343
    +0.12
    0.04
    76
    +0.11
    0.02
    Negative Logits
    ij
    -4.29
    ±
    -4.12
    Ł
    -4.02
    Ŀ
    -3.88
    ī
    -3.78
    ŀ
    -3.75
    Ĩ
    -3.74
    ļ
    -3.72
    °
    -3.69
    «
    -3.68
    POSITIVE LOGITS
     duplicates
    1.61
     increments
    1.54
     overflow
    1.52
     orderly
    1.48
    iqu
    1.44
    essential
    1.44
    arner
    1.43
     jumps
    1.39
     completes
    1.38
    ariable
    1.38
    Activations Density 0.173%

    No Known Activations