© 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. 30054
    Prev
    Next
    INDEX
    Explanations

    information relating to security disclosures and vulnerabilities

    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
    48
    +0.15
    0.8%
    157
    +0.14
    0.8%
    436
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    240
    +0.15
    -0.01
    289
    +0.14
    0.01
    371
    +0.12
    0.03
    Negative Logits
     palate
    -1.69
    thon
    -1.58
     veins
    -1.53
    ittest
    -1.52
     favour
    -1.41
     sore
    -1.40
    Decided
    -1.39
     fortun
    -1.35
    reland
    -1.33
     Records
    -1.32
    POSITIVE LOGITS
     embodiments
    2.21
     discloses
    1.75
     behaviors
    1.71
     disclosure
    1.70
     disclosed
    1.68
     patents
    1.65
     technologies
    1.60
     thereof
    1.59
     inventions
    1.53
     apparatuses
    1.51
    Activations Density 0.309%

    No Known Activations