© 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. 12214
    Prev
    Next
    INDEX
    Explanations

    references to appreciating or acknowledging service or products

    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
    300
    +0.13
    0.7%
    164
    +0.12
    0.7%
    70
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    213
    +0.13
    0.04
    164
    +0.12
    0.03
    440
    +0.12
    0.01
    Negative Logits
    verage
    -1.52
    IFF
    -1.49
    inki
    -1.48
     Hep
    -1.47
     multin
    -1.45
     binomial
    -1.45
     hypertension
    -1.42
    GTH
    -1.41
    abouts
    -1.38
    unnumbered
    -1.38
    POSITIVE LOGITS
    raisal
    2.93
    ropri
    2.79
    arent
    2.78
    lications
    2.76
    liance
    2.71
    arel
    2.67
    rehens
    2.50
    arently
    2.44
    lique
    2.29
    ellate
    2.28
    Activations Density 0.182%

    No Known Activations