© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    3. Pythia-70M-Deduped
    4. MLP Post
    5. 0-MLP-SM
    6. 26991
    Prev
    Next
    INDEX
    Explanations

    specific technical terms related to triggering mechanisms

    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
    285
    +0.13
    0.7%
    71
    +0.13
    0.7%
    2
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    111
    +0.13
    0.02
    59
    +0.13
    0.01
    317
    +0.12
    0.01
    Negative Logits
    ķ
    -1.65
    akes
    -1.63
    coat
    -1.52
    ¯
    -1.49
     lining
    -1.46
    '].
    -1.45
    Ķ
    -1.44
      
    -1.44
    ľĵ
    -1.42
    Ħ
    -1.42
    POSITIVE LOGITS
     law
    1.56
    roup
    1.54
    isa
    1.54
    igent
    1.52
     schemes
    1.51
     sciences
    1.47
     respects
    1.46
     scheme
    1.46
    erals
    1.44
    !(
    1.43
    Activations Density 0.021%

    No Known Activations