© 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. 1-RES-SM
    11. 26441
    Prev
    Next
    INDEX
    Explanations

    references to phones and phone-related topics

    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/1-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.1.hook_resid_post
    Hook Layer
    1
    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
    494
    +0.15
    0.9%
    173
    +0.14
    0.9%
    466
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    494
    +0.15
    0.03
    173
    +0.14
    0.03
    150
    +0.11
    0.03
    Negative Logits
     hope
    -1.88
    ĥ½
    -1.85
    ¨
    -1.84
    Ĥ
    -1.77
    athered
    -1.73
    ĥ
    -1.73
    olate
    -1.64
    Ļª
    -1.64
    ©
    -1.64
     union
    -1.57
    POSITIVE LOGITS
    gap
    2.13
    shots
    2.05
    ché
    1.81
    gun
    1.80
    caster
    1.76
    jack
    1.76
    ström
    1.75
    quel
    1.71
    estial
    1.66
    bing
    1.65
    Activations Density 0.150%

    No Known Activations