© 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. 12347
    Prev
    Next
    INDEX
    Explanations

    specific programming syntax or commands related to data handling

    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
    271
    +0.30
    1.9%
    64
    +0.23
    1.5%
    21
    +0.13
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    64
    +0.30
    0.11
    271
    +0.23
    0.02
    162
    +0.13
    0.03
    Negative Logits
    asone
    -2.02
    iginally
    -1.75
    htra
    -1.74
    ctions
    -1.66
    isers
    -1.61
    atrix
    -1.59
    isation
    -1.58
     Hammond
    -1.57
    itarian
    -1.54
    icense
    -1.53
    POSITIVE LOGITS
    Ī
    5.49
    Ļ
    5.43
    ¨
    5.39
    ®
    5.38
    «
    5.38
    Ĵ
    5.37
    ¤
    5.36
    Ģ
    5.33
    ı
    5.30
    ¬
    5.30
    Activations Density 0.109%

    No Known Activations