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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 29775
    Prev
    Next
    INDEX
    Explanations

    terms related to organizational 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/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
    510
    +0.15
    0.9%
    175
    +0.13
    0.8%
    79
    +0.13
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    510
    +0.15
    0.04
    282
    +0.13
    0.03
    175
    +0.13
    0.01
    Negative Logits
    documentclass
    -1.64
    ģ
    -1.54
    elig
    -1.50
    sburg
    -1.50
    uses
    -1.50
    д
    -1.44
    manuel
    -1.41
    CNN
    -1.34
    ocene
    -1.31
    urop
    -1.30
    POSITIVE LOGITS
    itating
    1.57
    onne
    1.53
     finances
    1.52
     affairs
    1.50
    odb
    1.50
    aste
    1.50
    ement
    1.46
    oise
    1.42
    ially
    1.42
     hereto
    1.41
    Activations Density 0.262%

    No Known Activations