© 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. 10907
    Prev
    Next
    INDEX
    Explanations

    instances of the word "confusion" and its variations, indicating themes of uncertainty or misunderstanding

    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
    172
    +0.13
    0.7%
    271
    +0.12
    0.6%
    322
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    386
    +0.13
    0.01
    213
    +0.12
    0.02
    218
    +0.11
    0.02
    Negative Logits
    veh
    -1.87
    whole
    -1.81
    walks
    -1.79
    manship
    -1.69
    ively
    -1.62
    cules
    -1.58
    mos
    -1.55
    wards
    -1.48
    bone
    -1.47
     whole
    -1.46
    POSITIVE LOGITS
    irmat
    2.74
    irmed
    2.36
    licts
    2.21
    irm
    2.13
    irms
    2.11
    idence
    2.06
    lict
    2.04
    ı
    2.02
    idential
    2.02
    ¨
    2.01
    Activations Density 0.102%

    No Known Activations