© 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. 12360
    Prev
    Next
    INDEX
    Explanations

    phrases related to the act of presenting or being present 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
    335
    +0.13
    0.7%
    16
    +0.13
    0.7%
    75
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    112
    +0.13
    0.02
    148
    +0.13
    0.01
    16
    +0.12
    0.02
    Negative Logits
    ½
    -2.02
    ĥ½
    -1.73
    ¥
    -1.67
    ŀ
    -1.66
    ·¸
    -1.64
    ´
    -1.57
    Ļª
    -1.55
    ¼
    -1.48
    utes
    -1.46
    âĢŁ
    -1.42
    POSITIVE LOGITS
    ence
    1.99
    grounds
    1.84
     tense
    1.76
    orrow
    1.76
    ences
    1.70
    emente
    1.69
    aneously
    1.68
    rums
    1.60
    aneous
    1.55
    orf
    1.53
    Activations Density 0.036%

    No Known Activations