© 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. 24068
    Prev
    Next
    INDEX
    Explanations

    references to the Boost software library

    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
    23
    +0.16
    0.9%
    177
    +0.11
    0.6%
    359
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    177
    +0.16
    0.02
    339
    +0.11
    0.02
    74
    +0.11
    0.01
    Negative Logits
    inson
    -1.64
    akespe
    -1.56
    ÃŃst
    -1.52
     influences
    -1.52
    vier
    -1.50
    afx
    -1.48
    ncbi
    -1.43
    onies
    -1.43
    aires
    -1.41
    Inflater
    -1.41
    POSITIVE LOGITS
    fitting
    1.77
    cing
    1.75
     dispute
    1.62
    ring
    1.55
    ced
    1.51
     developing
    1.46
     briefly
    1.44
     Member
    1.44
    ationally
    1.43
     publicly
    1.40
    Activations Density 0.097%

    No Known Activations