© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    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. 10522
    Prev
    Next
    INDEX
    Explanations

    phrases indicating abundance or quantity

    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
    307
    +0.32
    1.9%
    306
    +0.13
    0.8%
    228
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    101
    +0.32
    0.05
    125
    +0.13
    0.08
    145
    +0.11
    0.05
    Negative Logits
    ·
    -3.21
    «
    -3.19
    ½
    -3.17
    ¼
    -3.00
    ĺ
    -2.99
    »
    -2.89
    ģ
    -2.87
    <|outofrange|>
    -2.86
    <|outofrange|>
    -2.86
    -2.86
    POSITIVE LOGITS
     ways
    2.87
     examples
    2.83
     things
    2.69
     kinds
    2.65
     possibilities
    2.63
     exceptions
    2.54
     instances
    2.53
     reasons
    2.45
     alternatives
    2.42
     advantages
    2.39
    Activations Density 0.717%

    No Known Activations