© 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 EvalsExportsinterp-engineNEWAPI 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. 14040
    Prev
    Next
    INDEX
    Explanations

    references to species, authorship, and scientific categorization

    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
    78
    +0.15
    0.9%
    466
    +0.11
    0.7%
    375
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    479
    +0.15
    0.01
    321
    +0.11
    0.01
    241
    +0.10
    0.01
    Negative Logits
    ľĵ
    -1.67
    scriptions
    -1.65
    times
    -1.52
    ¬
    -1.51
     classrooms
    -1.51
    ¿
    -1.44
     buttons
    -1.43
    blogs
    -1.34
     listeners
    -1.34
     windows
    -1.33
    POSITIVE LOGITS
    acea
    1.53
     of
    1.51
     (;
    1.41
     belonging
    1.40
     endemic
    1.39
    ὶ
    1.34
    αÏĥ
    1.33
    nel
    1.29
    ALE
    1.28
     (,
    1.27
    Activations Density 0.025%

    No Known Activations