© 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. 11310
    Prev
    Next
    INDEX
    Explanations

    phrases related to specific species and their classification

    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
    21
    +0.20
    1.1%
    171
    +0.11
    0.6%
    362
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    362
    +0.20
    0.01
    452
    +0.11
    0.01
    356
    +0.10
    0.02
    Negative Logits
    ULTS
    -1.66
    IVE
    -1.66
    ativity
    -1.62
    oliberal
    -1.61
    APDH
    -1.58
    lining
    -1.56
     directions
    -1.53
    ifications
    -1.49
     onto
    -1.49
    ches
    -1.48
    POSITIVE LOGITS
    Ķ
    2.38
    ¿½
    2.35
    Ĵ
    2.34
    ¸
    2.31
    ¼
    2.27
    ĻĤ
    2.26
    ´
    2.23
    IJ
    2.21
    ĺ
    2.17
     genus
    2.15
    Activations Density 0.034%

    No Known Activations