© 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. 25834
    Prev
    Next
    INDEX
    Explanations

    references to governmental or organizational entities and their actions

    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
    +0.19
    1.1%
    493
    +0.15
    0.8%
    392
    +0.14
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    493
    +0.19
    0.07
    392
    +0.15
    0.09
    403
    +0.14
    0.06
    Negative Logits
    «
    -3.09
    ¥
    -2.73
    ¦
    -2.73
    ª
    -2.57
    °
    -2.56
    Ĥ¬
    -2.56
    <|outofrange|>
    -2.52
                ↵
    -2.52
                  ↵
    -2.52
    ↵
    -2.52
    POSITIVE LOGITS
     –
    1.78
    ).](
    1.74
    xia
    1.72
    –
    1.57
    pile
    1.55
    odon
    1.48
     âĨĴ
    1.48
    acts
    1.47
    colour
    1.45
    Greek
    1.40
    Activations Density 2.965%

    No Known Activations