© 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. 26356
    Prev
    Next
    INDEX
    Explanations

    words related to academic fields or disciplines, particularly those involving study and analysis

    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
    +0.18
    1.1%
    270
    +0.11
    0.6%
    348
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    270
    +0.18
    0.02
    417
    +0.11
    0.02
    406
    +0.11
    0.01
    Negative Logits
     certain
    -1.92
     pristine
    -1.68
     arbitrary
    -1.66
     pure
    -1.63
     Pure
    -1.57
     native
    -1.53
     particular
    -1.53
     plain
    -1.50
    ureus
    -1.49
    oused
    -1.48
    POSITIVE LOGITS
    ĻĤ
    2.29
    izer
    1.95
    »¿
    1.81
    ised
    1.79
    iller
    1.78
    mente
    1.75
    isms
    1.74
    Inflater
    1.74
    istas
    1.71
    gorithm
    1.67
    Activations Density 0.058%

    No Known Activations