© 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. 25463
    Prev
    Next
    INDEX
    Explanations

    HTML list items and structure definitions

    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
    +0.52
    3.5%
    111
    +0.35
    2.4%
    64
    +0.13
    0.9%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    138
    +0.52
    0.06
    333
    +0.35
    0.05
    49
    +0.13
    0.05
    Negative Logits
    rifug
    -1.24
     wells
    -1.19
    OLOR
    -1.18
     limitations
    -1.17
     inspiration
    -1.15
    icture
    -1.15
     Snyder
    -1.14
     centrifug
    -1.12
    sembled
    -1.12
     shortcomings
    -1.10
    POSITIVE LOGITS
    į
    4.79
    ĻĤ
    4.67
    ¼
    4.59
    Ħ
    4.46
    ¢
    4.44
    ³
    4.29
    ¸
    4.27
    «
    4.26
    ģ
    4.25
    ©
    4.24
    Activations Density 0.063%

    No Known Activations