© 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. Residuals
    9. 2-RES-SM
    11. 28304
    Prev
    Next
    INDEX
    Explanations

    expressions and phrases that convey emotional states or experiences

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    ctigges/pythia-70m-deduped__res-sm_processed/2-res-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_resid_post
    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
    159
    +0.14
    0.8%
    219
    +0.11
    0.6%
    6
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    33
    +0.14
    0.17
    407
    +0.11
    0.13
    88
    +0.11
    0.14
    Negative Logits
    Īĺ
    -2.73
    į
    -2.44
    ĭ
    -2.36
    ¿½
    -2.28
    ¼
    -2.27
    ħ
    -2.25
    ¤
    -2.23
    ª
    -2.22
    Ł
    -2.19
    ¸
    -2.16
    POSITIVE LOGITS
    à°¿
    1.52
     circulation
    1.36
    inher
    1.35
    )—
    1.35
    riel
    1.34
    equation
    1.32
    corresponding
    1.32
     $.
    1.31
     possible
    1.28
    ).$
    1.26
    Activations Density 4.666%

    No Known Activations