© 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. 26756
    Prev
    Next
    INDEX
    Explanations

    instances of the word "said" and its variants in spoken or reported dialogue

    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
    +0.18
    1.0%
    112
    +0.15
    0.8%
    170
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    112
    +0.18
    0.07
    26
    +0.15
    0.03
    101
    +0.13
    0.02
    Negative Logits
    pass
    -1.67
     pass
    -1.56
    load
    -1.56
    iman
    -1.51
    igu
    -1.49
    lay
    -1.47
     speeding
    -1.46
    looking
    -1.45
     brakes
    -1.43
    pull
    -1.43
    POSITIVE LOGITS
    Ĥ
    1.67
    ĨĴ
    1.63
    ¾
    1.60
    ľĵ
    1.48
     Univ
    1.48
    ı
    1.47
    Ĩ
    1.47
    ļ
    1.46
     interest
    1.40
    aceae
    1.33
    Activations Density 0.356%

    No Known Activations