© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. MLP Post
    9. 0-MLP-SM
    11. 27773
    Prev
    Next
    INDEX
    Explanations

    instances of high numerical values or strong emphasis on certain metrics

    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/0-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.0.hook_mlp_out
    Hook Layer
    0
    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
    472
    +0.12
    0.7%
    281
    +0.12
    0.6%
    198
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    438
    +0.12
    0.40
    53
    +0.12
    0.42
    56
    +0.12
    0.11
    Negative Logits
    ettes
    -2.05
    etting
    -1.69
    ets
    -1.64
    aku
    -1.52
    ERS
    -1.50
     United
    -1.47
    une
    -1.46
     upcoming
    -1.42
    IRS
    -1.42
    oku
    -1.40
    POSITIVE LOGITS
    ClCompile
    2.02
     consequence
    1.72
    ferent
    1.70
     neut
    1.65
     respect
    1.61
    Cause
    1.54
    stit
    1.53
     \%
    1.52
     care
    1.50
    ============
    1.47
    Activations Density 4.670%

    No Known Activations