© 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. 4-RES-SM
    11. 14760
    Prev
    Next
    INDEX
    Explanations

    numerical values related to configurations and resource limits

    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/4-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.4.hook_resid_post
    Hook Layer
    4
    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
    21
    +0.17
    1.0%
    156
    +0.17
    0.9%
    78
    +0.15
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    238
    +0.17
    0.03
    475
    +0.17
    0.06
    255
    +0.15
    0.00
    Negative Logits
     well
    -1.59
     dressing
    -1.58
    ↵
    -1.58
    ↵↵↵
    -1.58
    -1.58
    -1.58
    ↵↵
    -1.58
    ↵↵
    -1.58
     ↵
    -1.58
    ↵
    -1.58
    POSITIVE LOGITS
    0000000
    2.19
    px
    2.18
    00000000
    2.13
     /*!<
    2.11
    99999999
    2.00
     /*
    2.00
    000000
    1.98
    0000
    1.94
    ULL
    1.90
    vh
    1.89
    Activations Density 0.427%

    No Known Activations