© 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. 1-RES-SM
    11. 25518
    Prev
    Next
    INDEX
    Explanations

    terms related to exclusivity or distinction

    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
    262
    +0.15
    0.8%
    48
    +0.11
    0.6%
    300
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    92
    +0.15
    0.04
    262
    +0.11
    0.05
    48
    +0.10
    0.04
    Negative Logits
    ":
    -1.50
    abouts
    -1.45
    ':
    -1.38
    tu
    -1.38
    around
    -1.37
    day
    -1.36
    "?
    -1.36
    abl
    -1.35
    kas
    -1.32
    heet
    -1.31
    POSITIVE LOGITS
    fficients
    1.95
    escent
    1.63
     albicans
    1.61
    ETHOD
    1.55
    ociation
    1.51
    ially
    1.51
    opausal
    1.50
     facie
    1.48
    RECT
    1.45
    acter
    1.44
    Activations Density 0.477%

    No Known Activations