© 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. 25808
    Prev
    Next
    INDEX
    Explanations

    complex mathematical and programming expressions

    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
    98
    +0.16
    0.9%
    117
    +0.12
    0.7%
    20
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    313
    +0.16
    0.03
    223
    +0.12
    0.03
    354
    +0.12
    0.04
    Negative Logits
    ĨĴ
    -1.57
     exhausted
    -1.46
    eval
    -1.44
     humili
    -1.40
    ?'
    -1.39
    ilated
    -1.36
    $$
    -1.36
     trenches
    -1.36
    ensive
    -1.35
     inflicted
    -1.35
    POSITIVE LOGITS
    ":{"
    1.78
     franchise
    1.71
    ":{
    1.60
     forum
    1.55
     firmware
    1.42
    anguage
    1.42
    atus
    1.40
    ":"
    1.38
     website
    1.37
    atz
    1.36
    Activations Density 0.651%

    No Known Activations