© 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. 26880
    Prev
    Next
    INDEX
    Explanations

    specific terms related to modifications or changes

    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
    29
    +0.13
    0.7%
    92
    +0.13
    0.7%
    393
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    393
    +0.13
    0.01
    511
    +0.13
    0.03
    29
    +0.12
    0.03
    Negative Logits
    ittal
    -1.81
    trim
    -1.69
    athi
    -1.58
    .$
    -1.56
    ooting
    -1.52
    veolar
    -1.52
    vering
    -1.51
    Nil
    -1.49
    ottu
    -1.48
    oot
    -1.47
    POSITIVE LOGITS
    ķ
    3.25
    ¾
    3.11
    ħ
    3.07
    ı
    3.06
    Ĭ
    3.03
    Ŀ
    2.98
    ¸
    2.94
    ®
    2.93
    ŀ
    2.91
    »
    2.91
    Activations Density 0.929%

    No Known Activations