© 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. 27002
    Prev
    Next
    INDEX
    Explanations

    repeated characters or symbols, particularly slashes

    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
    +0.22
    1.2%
    210
    +0.12
    0.7%
    294
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    210
    +0.22
    0.02
    452
    +0.12
    0.03
    79
    +0.11
    0.04
    Negative Logits
    enum
    -1.57
    ygen
    -1.57
    ivat
    -1.49
    :`
    -1.43
    toString
    -1.43
    ulous
    -1.40
    uber
    -1.40
    oding
    -1.40
    asma
    -1.37
    rowing
    -1.33
    POSITIVE LOGITS
    ł
    2.51
    Ļª
    2.44
    ¦
    2.42
    İ
    2.35
    IJ
    2.34
    ¾
    2.28
    ¿
    2.26
    »
    2.25
    ¢
    2.23
    ±
    2.23
    Activations Density 0.120%

    No Known Activations