© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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. 4-RES-SM
    11. 11952
    Prev
    Next
    INDEX
    Explanations

    technical terminology related to microscopy and data extraction methodologies

    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
    78
    +0.17
    0.9%
    198
    +0.14
    0.8%
    85
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    85
    +0.17
    0.02
    148
    +0.14
    0.02
    443
    +0.13
    0.03
    Negative Logits
     defaults
    -1.59
     debts
    -1.54
     retire
    -1.48
     Own
    -1.44
     etiology
    -1.41
     occupants
    -1.40
     reserves
    -1.38
     vacant
    -1.38
    itals
    -1.37
     unhealthy
    -1.36
    POSITIVE LOGITS
     spectroscopy
    2.62
     analysis
    2.50
     microscopy
    2.40
     assay
    2.38
    imetry
    2.38
     scans
    2.35
     spectrometry
    2.33
     scanning
    2.29
     imaging
    2.24
     scan
    2.20
    Activations Density 0.196%

    No Known Activations