© 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. 4-RES-SM
    11. 12110
    Prev
    Next
    INDEX
    Explanations

    terms related to service or assistance

    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
    21
    +0.22
    1.3%
    453
    +0.11
    0.6%
    450
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    218
    +0.22
    0.05
    21
    +0.11
    0.03
    470
    +0.10
    0.05
    Negative Logits
     inheritance
    -1.78
    ploid
    -1.55
     chromosomes
    -1.55
    ffect
    -1.54
     synthesis
    -1.52
     exposition
    -1.51
     fragmented
    -1.43
     summation
    -1.40
     brightness
    -1.37
    ween
    -1.35
    POSITIVE LOGITS
    ¿
    4.13
    ©
    3.99
    ¾
    3.73
    ¦
    3.73
    Ł
    3.56
    £
    3.45
    ¢
    3.44
    ¿½
    3.44
    ¨
    3.43
    ¥
    3.40
    Activations Density 0.517%

    No Known Activations