© 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. 25866
    Prev
    Next
    INDEX
    Explanations

    phrases related to completion and fulfillment 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/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
    348
    +0.11
    0.6%
    123
    +0.11
    0.6%
    262
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    262
    +0.11
    0.06
    417
    +0.11
    0.05
    348
    +0.11
    0.06
    Negative Logits
    mered
    -1.61
     convenience
    -1.49
     possible
    -1.47
    ted
    -1.45
    ko
    -1.44
    ftware
    -1.42
     today
    -1.39
    bral
    -1.36
    needed
    -1.35
    kl
    -1.34
    POSITIVE LOGITS
    Īĺ
    2.53
    ĥ
    2.33
    ĥ½
    2.25
    Į
    2.14
    ĵ
    2.10
    Ī
    2.04
    ĸ´
    2.03
    ¿
    2.01
    ¼
    1.99
    ¿½
    1.97
    Activations Density 0.796%

    No Known Activations