© 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. 25580
    Prev
    Next
    INDEX
    Explanations

    references to the human body and its parts

    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
    95
    +0.13
    0.7%
    121
    +0.11
    0.6%
    23
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    297
    +0.13
    0.03
    58
    +0.11
    0.03
    23
    +0.11
    0.01
    Negative Logits
    ©
    -3.75
    º
    -3.71
    Ł
    -3.67
    Ĭ
    -3.45
    Ī
    -3.44
    ·
    -3.43
    Ĵ
    -3.42
    ¥
    -3.39
    ¬
    -3.36
    ł
    -3.35
    POSITIVE LOGITS
     sake
    1.68
     mood
    1.55
    ()->
    1.44
    gered
    1.43
     background
    1.41
    gett
    1.39
    eson
    1.37
    gery
    1.36
    hire
    1.31
     his
    1.30
    Activations Density 0.298%

    No Known Activations