© 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. 26487
    Prev
    Next
    INDEX
    Explanations

    references to the concept of "mixed" 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
    376
    +0.14
    0.8%
    15
    +0.12
    0.7%
    452
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    15
    +0.14
    0.02
    195
    +0.12
    0.01
    452
    +0.11
    0.02
    Negative Logits
    pired
    -1.55
    osp
    -1.53
    ubicin
    -1.42
    omen
    -1.42
    )$-
    -1.40
    aused
    -1.34
    oved
    -1.33
    ut
    -1.32
    cción
    -1.31
    overed
    -1.31
    POSITIVE LOGITS
    dle
    1.48
    plic
    1.47
    ↵³³
    1.45
    vist
    1.42
    bach
    1.42
    quist
    1.40
    vie
    1.39
    heid
    1.37
    uitar
    1.33
    ouin
    1.33
    Activations Density 0.054%

    No Known Activations