© 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. 1-RES-SM
    6. 27170
    Prev
    Next
    INDEX
    Explanations

    the word "cannot."

    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/1-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.1.hook_resid_post
    Hook Layer
    1
    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
    23
    +0.13
    0.7%
    202
    +0.11
    0.6%
    445
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    202
    +0.13
    0.04
    449
    +0.11
    0.02
    358
    +0.11
    0.03
    Negative Logits
    inger
    -1.88
    à«ĩ
    -1.73
    à¯į
    -1.71
    Ŀ
    -1.69
    ி
    -1.66
    »¿
    -1.65
    eenth
    -1.58
    à¯
    -1.57
     años
    -1.55
    à¯ģ
    -1.54
    POSITIVE LOGITS
    otherwise
    1.71
     anymore
    1.65
     miss
    1.59
     yet
    1.54
     fit
    1.52
     duplicate
    1.51
    empty
    1.46
     otherwise
    1.46
     empty
    1.45
    amen
    1.45
    Activations Density 0.269%

    No Known Activations