© 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. 26412
    Prev
    Next
    INDEX
    Explanations

    words related to dangerous or hazardous conditions

    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
    +0.13
    0.7%
    362
    +0.12
    0.7%
    466
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    111
    +0.13
    0.06
    271
    +0.12
    0.02
    156
    +0.11
    0.04
    Negative Logits
     such
    -1.63
     odd
    -1.62
     arbitrary
    -1.58
     liter
    -1.55
     colour
    -1.52
     certain
    -1.51
     complicated
    -1.48
     naught
    -1.47
     volatile
    -1.46
     exact
    -1.45
    POSITIVE LOGITS
    ŀ
    3.89
    £
    3.89
    º
    3.71
    Ŀ
    3.69
    ĺ
    3.64
    Ĵ
    3.58
    ¢
    3.57
    »¿
    3.53
    ¿
    3.50
    Ĭ
    3.48
    Activations Density 0.276%

    No Known Activations