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

    Neuronpedia

    Jacobian LensNEW
    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. 24395
    Prev
    Next
    INDEX
    Explanations

    sentences that convey legal disclaimers or conditions regarding liability

    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
    393
    +0.14
    0.8%
    251
    +0.11
    0.6%
    74
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    29
    +0.14
    0.03
    60
    +0.11
    0.02
    208
    +0.11
    0.03
    Negative Logits
    '))
    -2.13
    '));
    -1.96
    googleapis
    -1.93
    >).
    -1.70
    '),
    -1.63
    ))$.
    -1.60
    >)
    -1.53
    ')->
    -1.52
    otide
    -1.51
    ')$
    -1.50
    POSITIVE LOGITS
    İ
    2.05
    ¦
    1.98
    ł
    1.89
    Ĥ¬
    1.83
    ¹
    1.83
    »
    1.83
    ±
    1.79
    ³
    1.74
    ¾
    1.74
    Ĭ
    1.72
    Activations Density 0.346%

    No Known Activations