© 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. 28156
    Prev
    Next
    INDEX
    Explanations

    terms related to restrictions or limitations

    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
    463
    +0.14
    0.8%
    264
    +0.11
    0.6%
    11
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    11
    +0.14
    0.02
    463
    +0.11
    0.01
    179
    +0.10
    0.01
    Negative Logits
    ĻĤ
    -3.83
    Ĥ¬
    -3.81
    Ļª
    -3.76
    ĨĴ
    -3.71
    ĵ
    -3.69
    ¶
    -3.59
    ±
    -3.58
    Ĥ
    -3.56
    ·¸
    -3.54
    ´
    -3.52
    POSITIVE LOGITS
    stone
    2.09
    itol
    2.07
    itals
    1.97
    notes
    1.77
    stones
    1.75
    anguage
    1.73
    holder
    1.73
    ulated
    1.70
    ib
    1.69
    illar
    1.64
    Activations Density 0.453%

    No Known Activations