© 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 EvalsExportsinterp-engineNEWAPI 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. 31034
    Prev
    Next
    INDEX
    Explanations

    phrases related to communication or statements made by individuals

    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
    358
    +0.13
    0.7%
    2
    +0.12
    0.7%
    108
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    2
    +0.13
    0.04
    113
    +0.12
    0.04
    263
    +0.11
    0.04
    Negative Logits
    ij
    -3.00
    ·¸
    -2.97
    ľĵ
    -2.90
    ¬
    -2.89
    »¿
    -2.79
    ¸
    -2.70
    ¾
    -2.68
    ¨
    -2.63
    IJ
    -2.61
    Ĥ¬
    -2.58
    POSITIVE LOGITS
     goodbye
    2.92
     hello
    2.12
     farewell
    2.10
     harmless
    1.90
     prayers
    1.85
     bye
    1.84
     nothing
    1.72
     something
    1.67
     yes
    1.67
    ings
    1.60
    Activations Density 0.165%

    No Known Activations