© 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. 27451
    Prev
    Next
    INDEX
    Explanations

    terms related to environmental issues and sustainability

    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
    369
    +0.20
    1.2%
    59
    +0.12
    0.7%
    43
    +0.11
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    36
    +0.20
    0.03
    43
    +0.12
    0.03
    59
    +0.11
    0.02
    Negative Logits
     its
    -2.04
    inally
    -1.93
    ursuant
    -1.59
     gentle
    -1.58
    ctic
    -1.55
    ĸ´
    -1.53
     ever
    -1.51
     hers
    -1.50
    «
    -1.49
    orously
    -1.48
    POSITIVE LOGITS
    leaf
    2.50
    wich
    2.16
    berg
    2.08
    grass
    2.07
    wood
    1.91
    wald
    1.90
    wall
    1.89
    holm
    1.83
    bean
    1.81
    hill
    1.78
    Activations Density 0.082%

    No Known Activations