© 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 EvalsExportsinterp-engineNEWAPI 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. 24270
    Prev
    Next
    INDEX
    Explanations

    instances of specific names or identifiers and their corresponding significance

    oai_token-act-pair · gpt-4o-miniTriggered 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
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    271
    +0.14
    0.8%
    97
    +0.12
    0.7%
    213
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    347
    +0.14
    0.08
    200
    +0.12
    0.08
    97
    +0.12
    0.05
    Negative Logits
     ourselves
    -1.61
    ynes
    -1.57
     recovery
    -1.45
     navigate
    -1.39
     ([@
    -1.28
    iday
    -1.26
     powered
    -1.26
     defense
    -1.24
     decline
    -1.24
     defence
    -1.23
    POSITIVE LOGITS
    jas
    1.75
     slack
    1.58
    grad
    1.54
     attention
    1.53
    лем
    1.53
     Papers
    1.46
     rib
    1.45
    ICATION
    1.45
     hats
    1.40
     adj
    1.39
    Activations Density 2.784%

    No Known Activations