© 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. 4-RES-SM
    11. 10948
    Prev
    Next
    INDEX
    Explanations

    associations and statistical measures related to health risks and outcomes

    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
    318
    +0.14
    0.8%
    78
    +0.14
    0.8%
    153
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    369
    +0.14
    0.01
    11
    +0.14
    0.03
    153
    +0.13
    0.05
    Negative Logits
     named
    -1.66
     remedies
    -1.56
    binant
    -1.55
     Genome
    -1.54
     arsenal
    -1.50
     dialogue
    -1.50
     yet
    -1.47
     entity
    -1.44
     Gateway
    -1.42
     introdu
    -1.41
    POSITIVE LOGITS
    max
    2.38
    ratio
    2.29
    mean
    2.22
    peak
    2.12
    measured
    2.10
    avg
    2.09
     estimates
    2.08
     values
    2.07
    ·
    2.04
    Average
    2.04
    Activations Density 0.398%

    No Known Activations