© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    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. Residuals
    9. 2-RES-SM
    11. 29626
    Prev
    Next
    INDEX
    Explanations

    XML version declarations and elements related to XML schemas

    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
    307
    +0.14
    0.8%
    143
    +0.11
    0.6%
    197
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    307
    +0.14
    0.01
    345
    +0.11
    0.01
    21
    +0.10
    0.01
    Negative Logits
    .’”
    -1.71
    ...](
    -1.64
     advantage
    -1.64
    .’
    -1.54
    ?’
    -1.52
     planned
    -1.51
    ’.
    -1.46
     researchers
    -1.42
     ‘
    -1.41
     disadvantage
    -1.33
    POSITIVE LOGITS
     0000000000000000000000000000000000
    1.87
    aho
    1.68
    KB
    1.60
    ullivan
    1.51
    rå
    1.49
    000000
    1.46
    kb
    1.45
    åı·
    1.44
    riterion
    1.40
    xffffffff
    1.37
    Activations Density 0.055%

    No Known Activations