© 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. 30264
    Prev
    Next
    INDEX
    Explanations

    phrases and terms associated with clinical or technical elements and characteristics

    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
    98
    +0.18
    1.1%
    10
    +0.18
    1.1%
    478
    +0.17
    1.0%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    10
    +0.18
    0.16
    291
    +0.18
    0.14
    150
    +0.17
    0.12
    Negative Logits
    "}](#
    -1.73
    upgreek
    -1.55
    CoV
    -1.52
    ensch
    -1.52
    liness
    -1.43
    cule
    -1.43
    wagen
    -1.42
     regards
    -1.41
    feas
    -1.39
    Äį
    -1.38
    POSITIVE LOGITS
    Ľ
    3.41
    ĸ
    3.29
    ĨĴ
    3.27
    Ĺ
    2.97
    ↵  âĢĥ
    2.93
    ↵↵
    2.93
    ↵
    2.93
    ↵↵
    2.93
    ↵
    2.93
    čč
    2.93
    Activations Density 1.094%

    No Known Activations