© 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. 11586
    Prev
    Next
    INDEX
    Explanations

    conceptual frameworks and structural principles in various contexts

    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
    25
    +0.15
    0.8%
    268
    +0.11
    0.6%
    55
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    401
    +0.15
    0.07
    184
    +0.11
    0.09
    229
    +0.10
    0.09
    Negative Logits
    ©
    -4.58
    į
    -4.45
    ¥
    -4.36
    İ
    -4.30
    ŀ
    -4.30
    ľĵ
    -4.20
    µ
    -4.17
    ³
    -4.17
    Į
    -4.15
    ¡
    -4.12
    POSITIVE LOGITS
     themselves
    3.81
    creen
    3.46
    heet
    3.10
    pace
    3.07
    hip
    2.87
    hift
    2.86
    ystem
    2.85
    mith
    2.83
    elves
    2.83
    erver
    2.72
    Activations Density 0.945%

    No Known Activations