© 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. 11043
    Prev
    Next
    INDEX
    Explanations

    mathematical expressions involving division and negative numbers

    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
    26
    +0.12
    0.7%
    157
    +0.11
    0.6%
    391
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    26
    +0.12
    0.01
    283
    +0.11
    0.01
    359
    +0.10
    0.01
    Negative Logits
    .'"
    -1.66
    '."
    -1.65
    ()</
    -1.51
    ';
    -1.44
    \'
    -1.42
    msgstr
    -1.41
    ↵↵ ↵
    -1.40
    ³³
    -1.39
    .’”
    -1.39
    >';
    -1.35
    POSITIVE LOGITS
    ĨĴ
    2.93
    ¥
    2.58
    ¾
    2.50
    ¿
    2.43
    Ĵ
    2.29
    ¹
    2.29
    Īĺ
    2.25
    <|outofrange|>
    2.19
    ↵
    2.19
    ↵
    2.19
    Activations Density 0.029%

    No Known Activations