© 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. 25117
    Prev
    Next
    INDEX
    Explanations

    references to the order and occurrence of items, particularly in lists or sequences

    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
    +0.17
    0.9%
    320
    +0.14
    0.8%
    259
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    82
    +0.17
    0.03
    420
    +0.14
    0.04
    179
    +0.11
    0.04
    Negative Logits
    enic
    -1.56
    documentclass
    -1.53
    onies
    -1.46
     anywhere
    -1.45
    ftware
    -1.44
     fails
    -1.43
     {#
    -1.43
    lier
    -1.39
    paces
    -1.39
    lications
    -1.37
    POSITIVE LOGITS
    Īĺ
    4.00
    ¿
    4.00
    »¿
    3.98
    ¶
    3.94
    ®
    3.90
    Ļª
    3.87
    ´
    3.84
    ĺ
    3.84
    ¾
    3.83
    ³
    3.81
    Activations Density 0.153%

    No Known Activations