© 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. 10700
    Prev
    Next
    INDEX
    Explanations

    terms related to systematic reviews and methodological development in research context

    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
    98
    +0.17
    1.0%
    34
    +0.15
    0.9%
    418
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    418
    +0.17
    -0.01
    404
    +0.15
    0.13
    455
    +0.12
    0.14
    Negative Logits
     pills
    -1.85
     lol
    -1.71
     pounds
    -1.68
     injured
    -1.58
     cigarettes
    -1.53
    ipation
    -1.48
     pellets
    -1.46
    cium
    -1.43
     wrong
    -1.41
    urname
    -1.39
    POSITIVE LOGITS
     collaborative
    2.01
     stakeholders
    1.97
     initiatives
    1.91
     agenda
    1.85
     aims
    1.84
     conceptual
    1.78
     strengthening
    1.77
     empowered
    1.76
     facilit
    1.76
     broadening
    1.71
    Activations Density 4.490%

    No Known Activations