© 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. 23153
    Prev
    Next
    INDEX
    Explanations

    contexts involving the concept of "filling" or "fulfillment."

    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
    257
    +0.16
    0.9%
    171
    +0.15
    0.9%
    191
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    299
    +0.16
    0.02
    236
    +0.15
    0.03
    191
    +0.13
    0.04
    Negative Logits
     anthem
    -1.95
    heer
    -1.62
    DTD
    -1.60
    ↵³³
    -1.57
    SEA
    -1.57
    chos
    -1.51
    ád
    -1.50
     notch
    -1.49
    assadors
    -1.49
    ovember
    -1.47
    POSITIVE LOGITS
    Ī
    1.78
    °
    1.64
     lasting
    1.57
     certain
    1.50
     similar
    1.48
     powers
    1.41
     plot
    1.38
     Scient
    1.37
    }$\
    1.34
     plots
    1.32
    Activations Density 0.715%

    No Known Activations