© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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. 11964
    Prev
    Next
    INDEX
    Explanations

    code and commands related to programming and system management

    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
    307
    +0.41
    2.5%
    240
    +0.13
    0.8%
    447
    +0.11
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    307
    +0.41
    0.19
    17
    +0.13
    0.04
    47
    +0.11
    0.08
    Negative Logits
     personally
    -1.55
     makeup
    -1.54
     methodology
    -1.54
     expertise
    -1.47
    lay
    -1.46
     LLP
    -1.43
     Literature
    -1.43
     similarities
    -1.42
    seys
    -1.42
     methodologies
    -1.41
    POSITIVE LOGITS
    ĨĴ
    4.44
      ↵
    4.39
    ↵
    4.39
    ↵
    4.39
    ↵↵
    4.39
    4.39
    4.39
    4.39
    ↵
    4.39
                  ↵
    4.39
    Activations Density 4.394%

    No Known Activations