© 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. 2-RES-SM
    11. 10421
    Prev
    Next
    INDEX
    Explanations

    documents related to formal announcements or reports

    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
    365
    +0.16
    0.9%
    278
    +0.14
    0.8%
    78
    +0.13
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    365
    +0.16
    0.09
    340
    +0.14
    0.06
    212
    +0.13
    0.08
    Negative Logits
    áĥĺ
    -1.85
    dden
    -1.81
    iden
    -1.61
    autre
    -1.60
     drained
    -1.60
    áĥĶ
    -1.59
    abi
    -1.55
    à©
    -1.55
    soever
    -1.53
    ço
    -1.50
    POSITIVE LOGITS
     maps
    1.83
     weekly
    1.74
     proposals
    1.68
     versions
    1.59
     reports
    1.59
     plans
    1.57
     copies
    1.50
     proposal
    1.49
     monthly
    1.49
     tracker
    1.48
    Activations Density 2.284%

    No Known Activations