© 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. 12256
    Prev
    Next
    INDEX
    Explanations

    the term "og" indicating interest in the concept of "og," potentially relating to social dynamics or group behavior

    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
    410
    +0.16
    0.9%
    6
    +0.15
    0.8%
    95
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    410
    +0.16
    0.03
    349
    +0.15
    0.03
    181
    +0.10
    0.03
    Negative Logits
     dozen
    -1.80
     reasons
    -1.77
     steps
    -1.49
    grown
    -1.48
     certain
    -1.48
     chambers
    -1.46
     quick
    -1.46
    cribing
    -1.45
     attention
    -1.44
    igner
    -1.44
    POSITIVE LOGITS
    ¹
    3.83
    ĥ½
    3.50
    ª
    3.46
    ¸
    3.45
    ¼
    3.45
    ¾
    3.43
    İ
    3.42
    ·
    3.40
    º
    3.29
    Ģ
    3.24
    Activations Density 0.148%

    No Known Activations