© 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
    2. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    3. Pythia-70M-Deduped
    4. Residuals
    5. 4-RES-SM
    6. 11844
    Prev
    Next
    INDEX
    Explanations

    keywords and elements related to technical specifications and features of products

    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
    111
    +0.53
    3.9%
    156
    +0.49
    3.7%
    478
    +0.11
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    156
    +0.53
    0.07
    111
    +0.49
    0.08
    197
    +0.11
    0.07
    Negative Logits
    ®
    -1.43
    Ļ
    -1.27
    Ł
    -1.27
    inals
    -1.26
    ł
    -1.23
     Territory
    -1.19
    thood
    -1.17
     collect
    -1.14
    oks
    -1.11
     batches
    -1.07
    POSITIVE LOGITS
    medsc
    1.57
    agra
    1.47
    pone
    1.46
    substack
    1.42
     supra
    1.36
    ubottu
    1.35
    smallcaps
    1.34
    typeof
    1.33
     ibid
    1.32
    bell
    1.30
    Activations Density 0.044%

    No Known Activations