© 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. 2-RES-SM
    11. 30138
    Prev
    Next
    INDEX
    Explanations

    phrases that express causation or effects related to various subjects and conditions

    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/2-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.2.hook_resid_post
    Hook Layer
    2
    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
    +0.21
    1.2%
    23
    +0.16
    0.9%
    186
    +0.15
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    310
    +0.21
    0.02
    245
    +0.16
    0.03
    238
    +0.15
    0.02
    Negative Logits
    icular
    -1.70
    urally
    -1.63
    lite
    -1.59
    gio
    -1.58
     housed
    -1.57
    nai
    -1.56
    usepackage
    -1.55
    ica
    -1.55
    builder
    -1.53
    utor
    -1.53
    POSITIVE LOGITS
    ĻĤ
    3.53
    Ļª
    3.46
    ¿
    3.16
    ĨĴ
    3.14
    Ł
    3.04
    Ĵ
    3.02
    ↵
    2.99
    2.99
    ↵
    2.99
    2.99
    Activations Density 0.123%

    No Known Activations