© 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. 0-MLP-SM
    11. 26970
    Prev
    Next
    INDEX
    Explanations

    instances of high-frequency sounds or specific bell-like references in the text

    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
    369
    +0.16
    0.9%
    172
    +0.14
    0.8%
    450
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    111
    +0.16
    0.03
    211
    +0.14
    0.02
    491
    +0.13
    0.02
    Negative Logits
    »¿
    -2.21
    ·¸
    -2.18
    ĭ
    -2.04
    ļ
    -2.04
    İ
    -2.00
    į
    -1.94
    ¸
    -1.89
    ĥ
    -1.87
    ¦
    -1.82
    Ķ
    -1.81
    POSITIVE LOGITS
    otypic
    1.58
    rot
    1.57
    asic
    1.51
    inin
    1.49
     interpreted
    1.48
    ume
    1.46
    otype
    1.45
    icago
    1.45
    inese
    1.44
    otypes
    1.41
    Activations Density 0.110%

    No Known Activations