© 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. 10571
    Prev
    Next
    INDEX
    Explanations

    occurrences of the word "multiple" in various contexts

    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
    +0.20
    1.1%
    221
    +0.14
    0.8%
    364
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    221
    +0.20
    0.02
    101
    +0.14
    0.02
    202
    +0.13
    0.01
    Negative Logits
    deal
    -1.84
    UTION
    -1.76
    vette
    -1.66
     pudding
    -1.62
    ounge
    -1.60
    vy
    -1.56
    Ñĥб
    -1.56
    itian
    -1.55
    an
    -1.53
    ctrine
    -1.53
    POSITIVE LOGITS
     sclerosis
    3.11
    Ŀ
    2.51
    ples
    2.39
    į
    2.35
     generations
    2.33
    ħ
    2.20
     copies
    2.19
     occasions
    2.18
     sources
    2.17
    ŀ
    2.13
    Activations Density 0.096%

    No Known Activations