© 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
    2. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    3. Pythia-70M-Deduped
    4. Residuals
    5. 2-RES-SM
    6. 10926
    Prev
    Next
    INDEX
    Explanations

    terms related to survival rates and prognosis in a medical context

    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
    +0.17
    1.0%
    5
    +0.14
    0.8%
    254
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    335
    +0.17
    0.01
    5
    +0.14
    0.01
    239
    +0.12
    0.02
    Negative Logits
    ½
    -1.90
    ¿
    -1.88
    ´
    -1.85
    ®
    -1.84
    ķ
    -1.83
    ¿½
    -1.78
    lish
    -1.70
    «
    -1.69
    ish
    -1.68
    uscular
    -1.67
    POSITIVE LOGITS
     scene
    1.98
     accru
    1.80
     outcome
    1.70
     benefits
    1.62
    arium
    1.57
    arial
    1.57
     centers
    1.54
     underway
    1.51
     accrued
    1.50
     incurred
    1.50
    Activations Density 0.075%

    No Known Activations