© 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. 10494
    Prev
    Next
    INDEX
    Explanations

    references to background information in texts

    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
    156
    +0.49
    3.3%
    111
    +0.40
    2.7%
    161
    +0.13
    0.9%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    156
    +0.49
    0.32
    71
    +0.40
    0.29
    111
    +0.13
    0.30
    Negative Logits
    geant
    -1.30
     Console
    -1.29
    ĥ½
    -1.24
    varn
    -1.20
    cules
    -1.20
    <|outofrange|>
    -1.19
    ↵
    -1.19
    <|outofrange|>
    -1.19
    ↵
    -1.19
    ↵↵
    -1.19
    POSITIVE LOGITS
    edly
    1.69
    terday
    1.51
    irectional
    1.32
     controversy
    1.32
    ======
    1.29
    erty
    1.28
    :{
    1.24
    =>
    1.23
     =>
    1.22
     investigation
    1.19
    Activations Density 0.864%

    No Known Activations