© 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. 1-RES-SM
    11. 27432
    Prev
    Next
    INDEX
    Explanations

    instances of phrases or terms that indicate specificity or detailed focus on particular subjects

    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/1-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.1.hook_resid_post
    Hook Layer
    1
    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
    156
    +0.24
    1.4%
    362
    +0.16
    0.9%
    303
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    362
    +0.24
    0.01
    270
    +0.16
    0.01
    391
    +0.12
    0.01
    Negative Logits
    ÅŁ
    -1.52
    rile
    -1.51
    kan
    -1.39
    cester
    -1.37
    isenberg
    -1.36
    hs
    -1.35
    git
    -1.33
     Khal
    -1.33
     lui
    -1.31
    tait
    -1.31
    POSITIVE LOGITS
    ity
    1.88
    ities
    1.80
     aims
    1.70
    Īĺ
    1.65
     endings
    1.61
     regions
    1.60
    Ŀ
    1.59
     formats
    1.56
    erals
    1.55
    Ł
    1.53
    Activations Density 0.023%

    No Known Activations