© 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. 11972
    Prev
    Next
    INDEX
    Explanations

    structured data or syntax within programming languages

    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
    156
    +0.35
    2.1%
    271
    +0.17
    1.0%
    307
    +0.15
    0.9%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    273
    +0.35
    0.03
    9
    +0.17
    0.05
    400
    +0.15
    0.02
    Negative Logits
    .’”
    -1.64
    .'"
    -1.54
    nier
    -1.48
    atically
    -1.44
    idone
    -1.44
    ftware
    -1.41
    .’
    -1.35
    ?’
    -1.32
    eda
    -1.32
    .[]{
    -1.31
    POSITIVE LOGITS
    ·¸
    5.42
    ²
    5.27
    Ĥ
    5.12
    ĥ½
    5.12
    ¤
    5.07
    ¶
    5.07
    Ĥ¬
    5.01
    ij
    5.00
    £
    4.97
    «
    4.96
    Activations Density 0.347%

    No Known Activations