© 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. 26218
    Prev
    Next
    INDEX
    Explanations

    occurrences of the word "fatal" 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
    156
    +0.20
    1.1%
    326
    +0.13
    0.7%
    148
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    148
    +0.20
    0.02
    452
    +0.13
    0.02
    326
    +0.13
    0.02
    Negative Logits
    separator
    -1.54
     theirs
    -1.52
     ours
    -1.49
    uffling
    -1.47
    heet
    -1.38
    noop
    -1.38
    eeper
    -1.37
    hen
    -1.34
    unnumbered
    -1.34
     Paper
    -1.33
    POSITIVE LOGITS
    ĻĤ
    3.58
    ¾
    3.48
    Ł
    3.12
    ĥ
    3.12
    ¨
    3.09
    ¼
    3.09
    ¬
    3.07
    ¿½
    2.99
    ģ
    2.93
    Ĭ
    2.90
    Activations Density 0.129%

    No Known Activations