© 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. 0-MLP-SM
    11. 30270
    Prev
    Next
    INDEX
    Explanations

    the infinitive form of verbs, particularly "to" followed by action words

    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/0-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.0.hook_mlp_out
    Hook Layer
    0
    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
    206
    +0.13
    0.7%
    120
    +0.12
    0.7%
    5
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    69
    +0.13
    0.04
    109
    +0.12
    0.03
    434
    +0.12
    0.03
    Negative Logits
    inux
    -1.90
    imum
    -1.77
     argu
    -1.57
    alg
    -1.53
    uginosa
    -1.52
      ↵
    -1.51
    itage
    -1.45
    unicip
    -1.45
    camp
    -1.43
     liner
    -1.43
    POSITIVE LOGITS
    ¿½
    2.02
    ı
    1.93
    ¢
    1.91
    ¼
    1.76
    Ń
    1.74
    ice
    1.71
    į
    1.68
    Ĩ
    1.65
    ĩ
    1.65
    µ
    1.65
    Activations Density 0.027%

    No Known Activations