© 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. 24376
    Prev
    Next
    INDEX
    Explanations

    references to emotional experiences or sentiments

    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
    465
    +0.13
    0.7%
    508
    +0.11
    0.6%
    48
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    508
    +0.13
    0.11
    470
    +0.11
    0.15
    2
    +0.11
    0.10
    Negative Logits
    ¿½
    -4.04
    ¡
    -3.70
    ¿
    -3.66
    į
    -3.39
    ¾
    -3.38
    »
    -3.37
    ½
    -3.36
    «
    -3.36
    ª
    -3.27
    Ĺ
    -3.19
    POSITIVE LOGITS
    imar
    1.61
    wards
    1.59
    enix
    1.55
    yel
    1.54
    ↵
    1.48
    quest
    1.47
    encode
    1.43
    emen
    1.41
    ridges
    1.40
    ivat
    1.39
    Activations Density 2.586%

    No Known Activations