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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 26124
    Prev
    Next
    INDEX
    Explanations

    introductory letters and capitalized abbreviations

    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
    469
    +0.16
    0.9%
    397
    +0.15
    0.8%
    125
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    125
    +0.16
    0.08
    469
    +0.15
    0.06
    415
    +0.13
    0.05
    Negative Logits
     promoter
    -1.67
    ERTYPE
    -1.66
    iser
    -1.54
    enant
    -1.53
    legate
    -1.51
    istine
    -1.46
    oqu
    -1.45
    ienn
    -1.45
    enue
    -1.44
    rina
    -1.42
    POSITIVE LOGITS
    /*!
    1.64
    places
    1.43
    /*
    1.39
     presently
    1.39
     descent
    1.37
    âĢŁ
    1.36
    dump
    1.30
    course
    1.28
     strokes
    1.27
    irs
    1.27
    Activations Density 0.613%

    No Known Activations