© 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
    2. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    3. Pythia-70M-Deduped
    4. Residuals
    5. 4-RES-SM
    6. 9311
    Prev
    Next
    INDEX
    Explanations

    terms related to risk factors and health-related dangers

    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
    +0.11
    0.6%
    280
    +0.11
    0.6%
    82
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    82
    +0.11
    0.03
    437
    +0.11
    0.03
    221
    +0.10
    0.02
    Negative Logits
    thon
    -1.92
     decorated
    -1.67
     honor
    -1.61
    ters
    -1.45
    astery
    -1.44
     styled
    -1.43
    ynaptic
    -1.43
     ache
    -1.42
     opener
    -1.41
     excellence
    -1.40
    POSITIVE LOGITS
    ¹
    2.94
     factors
    2.80
    Ļª
    2.70
    ¶
    2.60
    ĺ
    2.60
    ¼
    2.50
     stratification
    2.44
     Factors
    2.40
    ³
    2.40
    ĻĤ
    2.39
    Activations Density 0.169%

    No Known Activations