© 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. 4-RES-SM
    6. 11957
    Prev
    Next
    INDEX
    Explanations

    references to specific studies or cases in scientific literature

    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
    +0.24
    1.4%
    255
    +0.15
    0.9%
    78
    +0.14
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    203
    +0.24
    0.01
    255
    +0.15
    0.00
    312
    +0.14
    0.01
    Negative Logits
     course
    -1.34
     operator
    -1.32
    .+
    -1.30
     ↵↵
    -1.28
     restoration
    -1.28
    verts
    -1.27
     '.
    -1.26
     moderation
    -1.26
     inscription
    -1.25
    ateurs
    -1.25
    POSITIVE LOGITS
    ©
    3.46
    ¡
    3.42
    ±
    3.32
    ģ
    3.25
    Ļ
    3.21
    ª
    3.12
    §
    3.08
    ¼
    3.06
    ı
    3.02
    °
    3.02
    Activations Density 0.029%

    No Known Activations