© 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. 2-MLP-SM
    11. 22410
    Prev
    Next
    INDEX
    Explanations

    programming constructs and elements related to data handling or manipulation

    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/2-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.2.hook_mlp_out
    Hook Layer
    2
    Architecture
    standard
    Context Size
    128
    Dataset
    EleutherAI/the_pile_deduplicated
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Neuron Alignment
    Index
    Value
    % of L₁
    23
    +0.22
    1.3%
    455
    +0.13
    0.8%
    471
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    23
    +0.22
    0.08
    279
    +0.13
    0.08
    51
    +0.12
    0.09
    Negative Logits
    ¸
    -1.77
    ¹
    -1.76
    ·
    -1.69
    ³
    -1.67
    esian
    -1.59
    ·¸
    -1.53
    quist
    -1.52
    ©
    -1.52
    Ģ
    -1.50
     Parameters
    -1.50
    POSITIVE LOGITS
     approached
    1.64
     shy
    1.53
     trouble
    1.51
     happ
    1.49
     elderly
    1.48
    ãģªãģĦ
    1.33
    ,’
    1.31
     hardest
    1.31
    [$
    1.30
     swept
    1.29
    Activations Density 1.101%

    No Known Activations