© 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. 31246
    Prev
    Next
    INDEX
    Explanations

    patterns of numerical data, particularly focusing on various numerical formats and structures

    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
    159
    +0.16
    1.0%
    287
    +0.14
    0.8%
    282
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    358
    +0.16
    0.16
    282
    +0.14
    0.12
    274
    +0.12
    0.15
    Negative Logits
    keit
    -1.54
    ories
    -1.47
     traits
    -1.44
    ther
    -1.42
     moments
    -1.40
     okay
    -1.39
     stakes
    -1.38
     threads
    -1.33
     tant
    -1.31
     alright
    -1.30
    POSITIVE LOGITS
    ĥ½
    2.55
    1.96
    1.96
           ↵
    1.96
    <|padding|>
    1.96
    1.96
    ↵
    1.96
    ↵
    1.96
    ↵  ³³³
    1.96
    ↵↵
    1.96
    Activations Density 2.979%

    No Known Activations