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

    Neuronpedia

    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. 25429
    Prev
    Next
    INDEX
    Explanations

    phrases and terms related to programming and software development

    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
    +0.56
    4.5%
    111
    +0.54
    4.3%
    200
    +0.10
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    200
    +0.56
    0.02
    138
    +0.54
    0.03
    66
    +0.10
    0.02
    Negative Logits
    ĥ½
    -2.03
    ®
    -1.97
    Ĥ¬
    -1.78
    ¬
    -1.70
    ¾
    -1.65
    Ĺ
    -1.62
    Ľ
    -1.62
    ij
    -1.61
    Ģ
    -1.61
    ¢
    -1.60
    POSITIVE LOGITS
    reb
    1.21
    bro
    1.16
     fundament
    1.13
    hetti
    1.09
    argin
    1.09
     eighteenth
    1.06
     facts
    1.06
     fantasy
    1.06
     good
    1.05
     adulthood
    1.05
    Activations Density 0.053%

    No Known Activations