Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    3. Pythia-70M-Deduped
    4. MLP Post
    5. 0-MLP-SM
    6. 27305
    Prev
    Next
    INDEX
    Explanations

    references to judicial processes and decisions

    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
    412
    +0.13
    0.8%
    71
    +0.12
    0.7%
    37
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    365
    +0.13
    0.01
    378
    +0.12
    0.01
    412
    +0.10
    0.01
    Negative Logits
    tox
    -2.02
     peace
    -1.81
     preponderance
    -1.66
     sweat
    -1.56
    rogens
    -1.54
     homeless
    -1.51
     ?"
    -1.49
     ':
    -1.47
    rosine
    -1.46
     abused
    -1.45
    POSITIVE LOGITS
    ļ
    2.48
    ī
    2.40
    ¡
    2.25
    Ĵ
    2.10
    ģ
    2.07
    Ĩ
    2.06
    ¢
    2.06
    IJ
    2.03
    CRIPT
    1.93
    ¨
    1.93
    Activations Density 0.325%

    No Known Activations