© 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. 2-RES-SM
    11. 31967
    Prev
    Next
    INDEX
    Explanations

    negations or phrases indicating the absence of something

    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
    59
    +0.10
    0.6%
    205
    +0.10
    0.6%
    478
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    329
    +0.10
    0.01
    375
    +0.10
    0.01
    205
    +0.10
    0.01
    Negative Logits
    inos
    -1.39
    aser
    -1.36
    oner
    -1.35
    hist
    -1.32
    noreply
    -1.29
    Observer
    -1.28
    send
    -1.27
    nutrition
    -1.26
    rophe
    -1.25
    cyt
    -1.24
    POSITIVE LOGITS
    ´
    2.02
     Redistributions
    1.96
    ĥ½
    1.89
    ©
    1.82
    ³
    1.79
    ¬
    1.79
    ĺ
    1.79
    º
    1.76
    ¡
    1.67
    ¿
    1.66
    Activations Density 0.040%

    No Known Activations