© 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. 4-RES-SM
    11. 14965
    Prev
    Next
    INDEX
    Explanations

    sections related to publishing and literary works

    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/4-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.4.hook_resid_post
    Hook Layer
    4
    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
    111
    +0.62
    5.4%
    156
    +0.49
    4.2%
    271
    +0.19
    1.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    111
    +0.62
    0.72
    156
    +0.49
    0.57
    71
    +0.19
    0.54
    Negative Logits
    ↵
    -2.82
    ↵
    -2.82
    č↵
    -2.82
    č↵
    -2.82
    ↵↵
    -2.82
    ↵
    -2.82
    ↵
    -2.82
    ↵
    -2.82
    ↵↵
    -2.82
    ↵↵
    -2.82
    POSITIVE LOGITS
    rosis
    1.24
     concentration
    1.22
    agra
    1.22
     agent
    1.18
     quien
    1.15
     sunt
    1.12
    ancellor
    1.12
    aea
    1.11
    ater
    1.09
     daughter
    1.09
    Activations Density 1.678%

    No Known Activations