© 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. 12829
    Prev
    Next
    INDEX
    Explanations

    keywords related to software applications and functionality

    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
    +0.51
    3.6%
    111
    +0.40
    2.8%
    25
    +0.16
    1.1%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    205
    +0.51
    0.45
    419
    +0.40
    0.37
    40
    +0.16
    0.44
    Negative Logits
    ĥ½
    -5.25
    <|padding|>
    -5.12
    <|outofrange|>
    -5.12
    <|outofrange|>
    -5.12
    <|outofrange|>
    -5.12
    ↵
    -5.12
    -5.12
    -5.12
    <|outofrange|>
    -5.12
    -5.12
    POSITIVE LOGITS
    atory
    1.41
    edia
    1.37
    imedia
    1.35
    ifice
    1.31
    etable
    1.30
     them
    1.28
    thems
    1.24
    ://
    1.23
    ater
    1.21
    olia
    1.21
    Activations Density 1.126%

    No Known Activations