© 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. 24780
    Prev
    Next
    INDEX
    Explanations

    occurrences of the same punctuation mark in various contexts

    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
    +0.16
    0.9%
    39
    +0.13
    0.7%
    104
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    269
    +0.16
    0.03
    184
    +0.13
    0.03
    474
    +0.13
    0.03
    Negative Logits
    ()).
    -1.60
    ()),
    -1.55
    hemat
    -1.54
    cession
    -1.50
    ÏĪ
    -1.49
    unctive
    -1.42
    >",
    -1.35
    arynge
    -1.34
    )",
    -1.32
    ())
    -1.31
    POSITIVE LOGITS
    ĥ
    2.42
    İ
    2.31
    ħ
    2.22
    ¿
    2.20
    ¾
    2.17
    ³
    2.08
    ·¸
    2.06
    ģ
    2.01
    Ĥ
    1.93
    ¼
    1.91
    Activations Density 0.099%

    No Known Activations