© 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. 12527
    Prev
    Next
    INDEX
    Explanations

    terms and phrases related to health conditions and their epidemiological factors

    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
    156
    +0.21
    1.2%
    153
    +0.16
    0.9%
    98
    +0.15
    0.9%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    261
    +0.21
    0.03
    98
    +0.16
    0.00
    489
    +0.15
    0.02
    Negative Logits
     nomination
    -1.75
     skills
    -1.69
     literacy
    -1.51
    !”
    -1.48
     hooks
    -1.48
     qualities
    -1.44
    '];
    -1.42
    ?",
    -1.41
     qualifications
    -1.40
     nets
    -1.35
    POSITIVE LOGITS
    ĻĤ
    3.21
    Īĺ
    3.00
    į
    2.98
    ł
    2.98
    ¦
    2.95
    ·¸
    2.94
    ģ
    2.93
    Ĵ
    2.91
    ĭ
    2.86
    ĸ´
    2.83
    Activations Density 0.519%

    No Known Activations