© 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. Residuals
    9. 1-RES-SM
    11. 26538
    Prev
    Next
    INDEX
    Explanations

    words indicating a serious or grave nature of situations or conditions

    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/1-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.1.hook_resid_post
    Hook Layer
    1
    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
    156
    +0.20
    1.1%
    334
    +0.13
    0.8%
    376
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    111
    +0.20
    0.05
    156
    +0.13
    0.04
    334
    +0.12
    0.04
    Negative Logits
    orth
    -1.46
     perfect
    -1.45
     exclusive
    -1.39
    bara
    -1.38
    rien
    -1.38
     Beautiful
    -1.37
    ders
    -1.37
    ene
    -1.36
    orte
    -1.35
     menos
    -1.35
    POSITIVE LOGITS
    ĻĤ
    2.93
    »¿
    2.79
    ĩ
    2.77
    ľĵ
    2.71
    ³
    2.63
    ĸ
    2.60
    ·
    2.51
    ¼
    2.41
    Ħ
    2.40
    ²
    2.40
    Activations Density 0.383%

    No Known Activations