© 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. 24340
    Prev
    Next
    INDEX
    Explanations

    elements of legal or formal documentation and structure

    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
    156
    +0.38
    2.3%
    130
    +0.11
    0.6%
    176
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    71
    +0.38
    0.79
    111
    +0.11
    0.76
    47
    +0.10
    0.76
    Negative Logits
     conjecture
    -1.63
    NER
    -1.42
    ,''
    -1.38
     paradox
    -1.32
     wisdom
    -1.31
    ableness
    -1.28
     wor
    -1.28
     exercise
    -1.27
     yourselves
    -1.27
    atore
    -1.26
    POSITIVE LOGITS
    (".
    1.61
    Inflater
    1.58
    matic
    1.52
    ::
    1.47
    ("#
    1.43
    illin
    1.40
    live
    1.37
    Enumerable
    1.35
    lections
    1.32
    zona
    1.32
    Activations Density 3.615%

    No Known Activations