© 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
    2. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    3. Pythia-70M-Deduped
    4. MLP Post
    5. 0-MLP-SM
    6. 27723
    Prev
    Next
    INDEX
    Explanations

    references to statistical percentages and figures

    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
    206
    +0.11
    0.6%
    479
    +0.11
    0.6%
    59
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    479
    +0.11
    0.04
    511
    +0.11
    0.04
    206
    +0.11
    0.04
    Negative Logits
    Inflater
    -1.75
    APTER
    -1.56
    WM
    -1.55
    sts
    -1.53
    ESULT
    -1.51
    untime
    -1.47
    ETHOD
    -1.46
     METHODS
    -1.45
     SERVICES
    -1.41
    mt
    -1.40
    POSITIVE LOGITS
    ois
    1.72
    enty
    1.71
     talks
    1.64
    imet
    1.58
     qué
    1.47
     thereof
    1.39
     quarters
    1.38
    ière
    1.37
     cuts
    1.37
     rose
    1.37
    Activations Density 3.169%

    No Known Activations