© 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 EvalsExportsinterp-engineNEWAPI 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. 13859
    Prev
    Next
    INDEX
    Explanations

    terms related to population and demographic statistics

    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
    335
    +0.14
    0.8%
    156
    +0.13
    0.8%
    240
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    335
    +0.14
    0.02
    206
    +0.13
    0.01
    196
    +0.12
    0.01
    Negative Logits
    º
    -4.36
    ĥ½
    -4.06
    ³
    -3.88
    ĻĤ
    -3.85
    <|outofrange|>
    -3.83
    ↵
    -3.83
    -3.83
    ↵
    -3.83
    <|outofrange|>
    -3.83
    <|outofrange|>
    -3.83
    POSITIVE LOGITS
    iles
    1.80
     estimates
    1.72
     estimate
    1.71
    views
    1.67
    iscus
    1.62
     controls
    1.61
     ratio
    1.56
     sampling
    1.56
    IMAGE
    1.56
    haus
    1.54
    Activations Density 0.093%

    No Known Activations