© 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. 1-RES-SM
    11. 27933
    Prev
    Next
    INDEX
    Explanations

    references to stack and memory management

    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/1-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.1.hook_resid_post
    Hook Layer
    1
    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
    +0.22
    1.3%
    250
    +0.14
    0.8%
    255
    +0.13
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    255
    +0.22
    0.02
    250
    +0.14
    0.03
    272
    +0.13
    0.02
    Negative Logits
    woke
    -1.65
     pregnant
    -1.64
     sick
    -1.60
     cystic
    -1.55
     someone
    -1.49
    ubicin
    -1.45
    ms
    -1.43
     whistle
    -1.42
     consent
    -1.42
     its
    -1.41
    POSITIVE LOGITS
    holder
    1.97
    overflow
    1.88
    iewicz
    1.81
    ings
    1.79
    Overflow
    1.74
    Panel
    1.73
    ript
    1.70
    bank
    1.61
    Definitions
    1.61
    ups
    1.60
    Activations Density 0.126%

    No Known Activations