© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. Residuals
    9. 4-RES-SM
    11. 14962
    Prev
    Next
    INDEX
    Explanations

    negative expressions or sentiments

    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/4-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.4.hook_resid_post
    Hook Layer
    4
    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
    78
    +0.14
    0.8%
    240
    +0.12
    0.7%
    282
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    275
    +0.14
    0.02
    97
    +0.12
    0.03
    200
    +0.11
    0.03
    Negative Logits
    ceral
    -1.65
    abolism
    -1.58
    LLY
    -1.56
    asio
    -1.54
    uate
    -1.42
    alli
    -1.39
    arians
    -1.36
    abase
    -1.34
    culate
    -1.31
    ines
    -1.31
    POSITIVE LOGITS
    Ķ
    2.63
    Ħ
    2.45
    ĭ
    2.39
    ħ
    2.35
    ĥ
    2.34
    ª
    2.32
    Ĥ
    2.32
    ¶
    2.31
    ¸
    2.27
    °
    2.27
    Activations Density 0.057%

    No Known Activations