© 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. 26675
    Prev
    Next
    INDEX
    Explanations

    permissions and conditions for software usage

    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
    294
    +0.12
    0.7%
    342
    +0.12
    0.6%
    157
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    253
    +0.12
    0.02
    452
    +0.12
    0.01
    285
    +0.11
    0.02
    Negative Logits
     contrary
    -1.61
     jeopardy
    -1.58
    "'
    -1.56
    amy
    -1.54
    aintiff
    -1.46
    pled
    -1.46
     rumors
    -1.45
    yet
    -1.43
    iful
    -1.43
    vere
    -1.41
    POSITIVE LOGITS
    ©
    1.75
    ļ
    1.72
    ĩ
    1.71
    IJ
    1.68
    Ĵ
    1.68
    ª
    1.68
    §
    1.64
    Ń
    1.61
    ¨
    1.57
    Īĺ
    1.53
    Activations Density 0.130%

    No Known Activations