© 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. 24302
    Prev
    Next
    INDEX
    Explanations

    technical terms and concepts related to analysis and methodologies in scientific papers

    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
    98
    +0.16
    0.9%
    80
    +0.13
    0.7%
    500
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    500
    +0.16
    0.03
    155
    +0.13
    0.05
    266
    +0.13
    0.01
    Negative Logits
     perfect
    -1.71
     prejudice
    -1.67
     pains
    -1.60
     envy
    -1.59
     intuition
    -1.59
     favor
    -1.54
     advance
    -1.53
     &\
    -1.50
     ignorance
    -1.50
     evident
    -1.48
    POSITIVE LOGITS
    namespace
    1.64
    ®
    1.64
    ĥ½
    1.62
    »
    1.58
    Accessor
    1.54
    Registration
    1.52
     Register
    1.50
    <|outofrange|>
    1.49
    <|outofrange|>
    1.49
    1.49
    Activations Density 0.324%

    No Known Activations