© 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. 2-RES-SM
    11. 12359
    Prev
    Next
    INDEX
    Explanations

    key phrases related to crafting and DIY activities

    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
    465
    +0.16
    0.9%
    322
    +0.12
    0.6%
    222
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    465
    +0.16
    0.14
    407
    +0.12
    0.09
    404
    +0.11
    0.09
    Negative Logits
    ĻĤ
    -3.88
    »
    -3.75
    µ
    -3.62
    Ń
    -3.44
    ģ
    -3.41
    »¿
    -3.40
    ħ
    -3.37
    Ĥ
    -3.37
    Ĥ¬
    -3.36
    ¬
    -3.32
    POSITIVE LOGITS
     meant
    1.55
    ogn
    1.49
    ```
    1.49
     boolean
    1.48
     imperative
    1.48
    ++)
    1.47
    completion
    1.46
    optional
    1.45
     tedious
    1.44
    gradle
    1.43
    Activations Density 3.103%

    No Known Activations