© 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
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. Residuals
    9. 2-RES-SM
    11. 28466
    Prev
    Next
    INDEX
    Explanations

    the concept of separation and distinctiveness in various contexts

    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
    387
    +0.14
    0.8%
    239
    +0.12
    0.7%
    359
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    101
    +0.14
    0.01
    387
    +0.12
    0.02
    359
    +0.11
    0.02
    Negative Logits
    ines
    -1.69
    Ø°
    -1.45
    emis
    -1.44
    ophyll
    -1.42
    fter
    -1.39
    áĥĺáĥ
    -1.36
    Ùħ
    -1.36
    ARD
    -1.36
    ds
    -1.36
    ellow
    -1.35
    POSITIVE LOGITS
     lines
    1.93
    ively
    1.84
     generations
    1.75
     ones
    1.74
     sexes
    1.71
     than
    1.65
    ?).
    1.65
    iveness
    1.65
     passages
    1.54
     configurations
    1.50
    Activations Density 0.090%

    No Known Activations