© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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. 26921
    Prev
    Next
    INDEX
    Explanations

    phrases related to the concept of physical actions or behaviors

    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
    156
    +0.25
    1.4%
    21
    +0.11
    0.6%
    374
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    170
    +0.25
    0.04
    33
    +0.11
    0.04
    328
    +0.10
    0.05
    Negative Logits
    yss
    -1.66
    igt
    -1.64
    estock
    -1.63
    vers
    -1.61
    ich
    -1.61
    igan
    -1.60
    ï¬
    -1.58
    esses
    -1.56
    icol
    -1.53
    igh
    -1.48
    POSITIVE LOGITS
     fertile
    1.47
    rifug
    1.44
     upload
    1.43
     citizenship
    1.43
    ļ
    1.39
    à°¿
    1.38
     marriages
    1.38
     copyrighted
    1.37
     feature
    1.35
     eligible
    1.35
    Activations Density 0.362%

    No Known Activations