Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    3. Pythia-70M-Deduped
    4. Residuals
    5. 2-RES-SM
    6. 12019
    Prev
    Next
    INDEX
    Explanations

    Licensing terms and conditions related to software

    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
    192
    +0.17
    0.9%
    103
    +0.10
    0.6%
    319
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    192
    +0.17
    0.06
    116
    +0.10
    0.05
    103
    +0.10
    0.05
    Negative Logits
     relief
    -1.68
    kes
    -1.43
     prem
    -1.42
    imals
    -1.40
    grid
    -1.39
     installation
    -1.38
    ays
    -1.33
    ani
    -1.33
     percentage
    -1.32
     aisle
    -1.32
    POSITIVE LOGITS
    sible
    1.79
    ане
    1.66
    igator
    1.57
    алÑĮ
    1.55
    ----------------------------------------------------
    1.54
    ------------------------------------------------
    1.50
     adv
    1.48
    ainer
    1.48
    енÑĮ
    1.47
    Ñīе
    1.45
    Activations Density 0.086%

    No Known Activations