© 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. 12802
    Prev
    Next
    INDEX
    Explanations

    references to careers and professional paths

    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
    72
    +0.15
    0.8%
    172
    +0.13
    0.8%
    176
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    72
    +0.15
    0.02
    501
    +0.13
    0.01
    288
    +0.12
    0.03
    Negative Logits
    ing
    -2.21
    à°¿
    -1.73
    iang
    -1.72
    àµį
    -1.70
    idge
    -1.65
    à«ĩ
    -1.63
    ingen
    -1.62
    àµ
    -1.56
    heid
    -1.53
    cels
    -1.52
    POSITIVE LOGITS
     spanning
    2.28
     path
    2.09
     paths
    1.90
     expectancy
    1.88
     prospects
    1.88
     spans
    1.71
     aside
    1.66
     advancement
    1.66
     ending
    1.65
     spanned
    1.64
    Activations Density 0.160%

    No Known Activations