© 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. Residuals
    5. 2-RES-SM
    6. 28241
    Prev
    Next
    INDEX
    Explanations

    elements that define licensing and renewal details

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    ctigges/pythia-70m-deduped__res-sm_processed/2-res-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_resid_post
    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
    231
    +0.15
    0.9%
    271
    +0.14
    0.8%
    12
    +0.14
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    72
    +0.15
    0.11
    447
    +0.14
    0.04
    231
    +0.14
    0.15
    Negative Logits
    wagen
    -1.60
    creas
    -1.47
    ignon
    -1.38
    icken
    -1.37
    header
    -1.34
     next
    -1.33
     sunny
    -1.31
     minute
    -1.30
    noreply
    -1.28
     alright
    -1.27
    POSITIVE LOGITS
    onica
    1.45
    wordpress
    1.37
     society
    1.36
     coma
    1.34
    ichte
    1.33
    Verlag
    1.33
     chloride
    1.32
    aceae
    1.31
    à½
    1.29
    ourier
    1.29
    Activations Density 8.003%

    No Known Activations