© 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. 26821
    Prev
    Next
    INDEX
    Explanations

    various forms of parentheses used in the text

    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
    170
    +0.13
    0.7%
    130
    +0.12
    0.7%
    369
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    489
    +0.13
    0.02
    323
    +0.12
    0.02
    268
    +0.10
    0.03
    Negative Logits
    »¿
    -2.33
    ĻĤ
    -2.17
    ĵ
    -2.13
    ¿½
    -2.06
    ¼
    -1.93
    ľĵ
    -1.93
    ½
    -1.86
    ©
    -1.84
    Ń
    -1.83
    ł
    -1.81
    POSITIVE LOGITS
    hn
    1.46
    eded
    1.43
    !)
    1.42
     elapsed
    1.42
     briefs
    1.35
    ymp
    1.32
    ↵
    1.31
     soon
    1.31
     Appellate
    1.28
     Appeals
    1.27
    Activations Density 0.071%

    No Known Activations