© 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. 12417
    Prev
    Next
    INDEX
    Explanations

    references to figures or illustrations in the text

    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
    490
    +0.11
    0.6%
    410
    +0.11
    0.6%
    26
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    38
    +0.11
    0.01
    386
    +0.11
    0.01
    221
    +0.10
    0.01
    Negative Logits
    ish
    -1.49
     authorities
    -1.40
    issue
    -1.38
    EventListener
    -1.38
     spectrometer
    -1.35
    ctic
    -1.33
    ember
    -1.33
    à±į
    -1.30
    ict
    -1.30
    Created
    -1.30
    POSITIVE LOGITS
    ±
    1.96
    ª
    1.79
    ĥ
    1.73
    URES
    1.63
    Ń
    1.63
    tica
    1.62
    Ĭ
    1.55
    ģ
    1.55
    ħ
    1.55
    §
    1.55
    Activations Density 0.121%

    No Known Activations