© 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. 22533
    Prev
    Next
    INDEX
    Explanations

    expressions of aspiration or desire for positive outcomes

    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
    318
    +0.11
    0.6%
    452
    +0.10
    0.6%
    511
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    318
    +0.11
    0.01
    452
    +0.10
    0.01
    414
    +0.10
    0.01
    Negative Logits
    ugu
    -1.73
    romycin
    -1.63
    iph
    -1.54
    orer
    -1.53
    ursuant
    -1.51
    ibraries
    -1.49
    riber
    -1.46
    ftware
    -1.46
    opyright
    -1.44
    enix
    -1.44
    POSITIVE LOGITS
    ©
    2.67
    ¾
    2.41
    ĻĤ
    2.32
    ³
    2.23
    ķ
    2.19
    ¨
    2.17
    Ĭ
    2.15
    ¤
    2.14
    ļ
    2.09
    ¦
    2.09
    Activations Density 0.037%

    No Known Activations