© 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. 4-RES-SM
    11. 13077
    Prev
    Next
    INDEX
    Explanations

    phrases indicating software licensing or copyright information

    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
    21
    +0.22
    1.3%
    307
    +0.12
    0.7%
    242
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    242
    +0.22
    0.03
    98
    +0.12
    0.01
    195
    +0.11
    0.03
    Negative Logits
    atation
    -1.57
    aliana
    -1.55
     succession
    -1.52
    plasia
    -1.51
    vance
    -1.49
    ationally
    -1.49
    uality
    -1.49
    wright
    -1.47
     turnover
    -1.47
     thereafter
    -1.46
    POSITIVE LOGITS
     libc
    2.13
     lib
    2.10
     GNU
    2.04
     COPYRIGHT
    1.92
     GPL
    1.84
    libc
    1.81
     Linux
    1.81
    mozilla
    1.72
     gcc
    1.72
     org
    1.70
    Activations Density 0.180%

    No Known Activations