© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 29790
    Prev
    Next
    INDEX
    Explanations

    biochemical markers and substances related to Alzheimer's disease

    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
    271
    +0.23
    1.4%
    237
    +0.19
    1.2%
    161
    +0.16
    1.0%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    111
    +0.23
    0.33
    156
    +0.19
    0.33
    71
    +0.16
    0.31
    Negative Logits
    veis
    -1.67
    FU
    -1.48
     timeline
    -1.33
    cknowled
    -1.31
    amar
    -1.30
     pestic
    -1.30
     Conclusions
    -1.30
    ossier
    -1.30
     tele
    -1.28
     manifest
    -1.27
    POSITIVE LOGITS
    Äģr
    1.58
     issue
    1.48
    )',
    1.45
    istor
    1.43
    hee
    1.41
     feat
    1.37
    hell
    1.37
    pmod
    1.36
    rown
    1.36
     thrown
    1.36
    Activations Density 5.272%

    No Known Activations