© 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 EvalsExportsAPI 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. 25029
    Prev
    Next
    INDEX
    Explanations

    references to people involved in social or transactional 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/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
    252
    +0.12
    0.7%
    400
    +0.12
    0.7%
    111
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    2
    +0.12
    0.07
    100
    +0.12
    0.09
    88
    +0.11
    0.05
    Negative Logits
     himself
    -1.69
     herself
    -1.69
    itudes
    -1.59
    emente
    -1.50
    ment
    -1.43
     ceiling
    -1.43
     standards
    -1.41
     itself
    -1.41
    ements
    -1.38
    tees
    -1.35
    POSITIVE LOGITS
    generate
    1.46
    ourced
    1.45
    ãģĤãĤĭ
    1.44
    addEventListener
    1.42
    ätt
    1.39
    ="${
    1.37
    ognition
    1.36
    ADVERTISEMENT
    1.34
     aim
    1.33
    à±į
    1.33
    Activations Density 1.116%

    No Known Activations