© 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. 22543
    Prev
    Next
    INDEX
    Explanations

    the term "follow-up" in various contexts

    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
    31
    +0.11
    0.6%
    419
    +0.11
    0.6%
    273
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    287
    +0.11
    0.03
    88
    +0.11
    0.02
    281
    +0.11
    0.01
    Negative Logits
    ĻĤ
    -3.07
    ¦
    -2.61
    ¤
    -2.58
    ĺ
    -2.55
    ¥
    -2.54
    ¸
    -2.54
    °
    -2.51
    ³
    -2.41
    Ļª
    -2.37
    ķ
    -2.23
    POSITIVE LOGITS
     themselves
    1.56
     plane
    1.52
     purpose
    1.47
     timely
    1.38
    hin
    1.36
    .|
    1.35
     sender
    1.33
     Convention
    1.32
     coincidence
    1.31
     fert
    1.30
    Activations Density 0.275%

    No Known Activations