© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    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. 10272
    Prev
    Next
    INDEX
    Explanations

    the word "like" used in various contexts

    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
    178
    +0.14
    0.8%
    81
    +0.13
    0.7%
    274
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    91
    +0.14
    0.06
    327
    +0.13
    0.06
    81
    +0.13
    0.05
    Negative Logits
    etermined
    -1.68
    oster
    -1.52
    THER
    -1.51
    iture
    -1.50
    erson
    -1.47
    Introduction
    -1.44
    >&
    -1.42
    orte
    -1.40
    ....
    -1.40
    itory
    -1.36
    POSITIVE LOGITS
    ĨĴ
    2.75
    ¼
    2.64
    ¿
    2.56
    Ķ
    2.48
    Į
    2.47
    ĸ
    2.42
    Ĵ
    2.40
     ↵
    2.37
              ↵
    2.37
    ↵
    2.37
    Activations Density 0.226%

    No Known Activations