© 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. 23444
    Prev
    Next
    INDEX
    Explanations

    repetitive conjunctions or phrases that suggest a connection between different ideas

    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
    274
    +0.14
    0.8%
    409
    +0.11
    0.6%
    500
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    274
    +0.14
    0.07
    208
    +0.11
    0.05
    409
    +0.11
    0.06
    Negative Logits
    rops
    -1.85
    kes
    -1.73
    ker
    -1.73
    ³
    -1.69
    ķ
    -1.61
    poons
    -1.60
    igh
    -1.60
    orphisms
    -1.59
    ¦
    -1.59
    ighs
    -1.58
    POSITIVE LOGITS
    Verlag
    1.96
     possession
    1.74
    ensa
    1.56
    ugu
    1.53
    ophone
    1.48
    ée
    1.48
    forge
    1.47
    ène
    1.47
     magazine
    1.44
    FIL
    1.43
    Activations Density 0.255%

    No Known Activations