© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. MLP Post
    9. 0-MLP-SM
    11. 24944
    Prev
    Next
    INDEX
    Explanations

    terms related to the state of being dried or desiccated

    oai_token-act-pair · gpt-4o-mini
    Triggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    ctigges/pythia-70m-deduped__mlp-sm_processed/0-mlp-sm
    Prompts (Dashboard)
    32,768 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.0.hook_mlp_out
    Hook Layer
    0
    Architecture
    standard
    Context Size
    128
    Dataset
    EleutherAI/the_pile_deduplicated
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Neuron Alignment
    Index
    Value
    % of L₁
    156
    +0.17
    1.0%
    376
    +0.14
    0.8%
    474
    +0.13
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    422
    +0.17
    0.01
    474
    +0.14
    0.01
    194
    +0.13
    0.01
    Negative Logits
    ONS
    -1.76
    ...](
    -1.66
    arer
    -1.56
    watson
    -1.50
    olen
    -1.48
    opic
    -1.46
    hes
    -1.44
    ")]
    -1.44
    uen
    -1.43
    aliana
    -1.42
    POSITIVE LOGITS
    rant
    1.85
    ong
    1.72
    illo
    1.67
     (\>
    1.64
     apart
    1.58
    itto
    1.57
     version
    1.51
    blood
    1.48
    jack
    1.47
     emergencies
    1.47
    Activations Density 0.012%

    No Known Activations