© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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. 12346
    Prev
    Next
    INDEX
    Explanations

    verbs related to research studies and medical investigations

    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
    15
    +0.13
    0.7%
    209
    +0.12
    0.7%
    499
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    15
    +0.13
    0.03
    209
    +0.12
    0.03
    150
    +0.11
    0.02
    Negative Logits
    Ļª
    -2.84
    »¿
    -2.82
    ĥ½
    -2.62
    ĸ
    -2.44
    ĩ
    -2.44
    §
    -2.43
    ¨
    -2.33
    Ĩ
    -2.28
    ¿½
    -2.23
    ¥
    -2.23
    POSITIVE LOGITS
     prospectively
    2.21
     lapar
    2.12
     independently
    2.09
     sequentially
    2.08
     prior
    2.05
     concurrently
    2.02
    pez
    2.01
     using
    1.99
     examining
    1.98
     simultaneously
    1.97
    Activations Density 0.054%

    No Known Activations