© 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. 11225
    Prev
    Next
    INDEX
    Explanations

    phrases related to cognitive processes and decision-making

    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
    240
    +0.16
    0.9%
    293
    +0.12
    0.7%
    409
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    16
    +0.16
    0.12
    33
    +0.12
    0.14
    409
    +0.12
    0.15
    Negative Logits
    Īĺ
    -2.23
    Ŀ
    -2.17
    ħ
    -2.06
    ŀ
    -1.99
    ¿½
    -1.98
    §
    -1.96
    ķ
    -1.95
    ĻĤ
    -1.90
    ľ
    -1.86
    ģ
    -1.80
    POSITIVE LOGITS
    rote
    1.42
    blogspot
    1.41
    dp
    1.40
    pa
    1.36
    ncbi
    1.35
    zig
    1.33
    ipro
    1.33
    ze
    1.30
    roc
    1.29
    olla
    1.28
    Activations Density 3.150%

    No Known Activations