© 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
    2. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    3. Pythia-70M-Deduped
    4. MLP Post
    5. 0-MLP-SM
    6. 27119
    Prev
    Next
    INDEX
    Explanations

    specific repeated characters or patterns in text

    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
    274
    +0.14
    0.8%
    113
    +0.12
    0.7%
    150
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    274
    +0.14
    0.02
    276
    +0.12
    0.02
    161
    +0.10
    0.02
    Negative Logits
    opause
    -1.64
    ¿½
    -1.63
    Į
    -1.56
    Īĺ
    -1.53
    terday
    -1.51
    »¿
    -1.51
    Ĥ
    -1.50
    Ļª
    -1.48
    abeth
    -1.44
    ĻĤ
    -1.43
    POSITIVE LOGITS
    hold
    1.73
    gon
    1.64
    venir
    1.63
    âĢij
    1.59
    esan
    1.56
    CHA
    1.55
    coin
    1.46
    holds
    1.45
    venue
    1.44
    gos
    1.41
    Activations Density 1.558%

    No Known Activations