© 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 EvalsExportsinterp-engineNEWAPI 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. 9136
    Prev
    Next
    INDEX
    Explanations

    references to phone communication and related terminology

    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
    263
    +0.14
    0.8%
    348
    +0.12
    0.7%
    44
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    164
    +0.14
    0.05
    263
    +0.12
    0.04
    189
    +0.10
    0.03
    Negative Logits
    ¨
    -3.43
    Ľ
    -3.35
    ĸ
    -3.33
    ´
    -3.27
    ±
    -3.25
    Ĥ
    -3.23
    ĵ
    -3.17
    ½
    -3.17
    Ń
    -3.13
    £
    -3.12
    POSITIVE LOGITS
    gap
    2.10
     conversations
    2.01
     calls
    1.95
     conversation
    1.95
     number
    1.89
     booth
    1.87
     ringing
    1.86
     towers
    1.85
     phone
    1.81
     wire
    1.80
    Activations Density 0.367%

    No Known Activations