© 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. 0-MLP-SM
    11. 26045
    Prev
    Next
    INDEX
    Explanations

    phrases that indicate improvement or enhancement in various contexts

    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/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
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    156
    +0.34
    2.1%
    376
    +0.22
    1.3%
    115
    +0.13
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    408
    +0.34
    0.02
    327
    +0.22
    0.01
    62
    +0.13
    0.01
    Negative Logits
    sel
    -1.68
    CUIT
    -1.61
    west
    -1.58
    UAL
    -1.58
    unnumbered
    -1.57
    ariant
    -1.53
    rock
    -1.52
    liche
    -1.50
    naire
    -1.49
    illian
    -1.46
    POSITIVE LOGITS
    ł
    1.67
     detection
    1.63
    ĥ
    1.44
     efficiency
    1.43
     radius
    1.41
    ivat
    1.40
     fields
    1.39
    ively
    1.38
     situations
    1.36
     velocity
    1.36
    Activations Density 0.012%

    No Known Activations