© 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. 11029
    Prev
    Next
    INDEX
    Explanations

    terms related to requests and requirements

    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
    154
    +0.12
    0.6%
    271
    +0.12
    0.6%
    364
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    154
    +0.12
    0.03
    371
    +0.12
    0.02
    364
    +0.11
    0.02
    Negative Logits
    ially
    -1.85
    odies
    -1.61
    itary
    -1.60
    astro
    -1.53
    aurus
    -1.53
     compositions
    -1.50
    emic
    -1.46
    uous
    -1.45
     returns
    -1.41
    entially
    -1.39
    POSITIVE LOGITS
    ³
    2.91
    ´
    2.88
    ¬
    2.84
    ·
    2.82
    ¿
    2.78
    ¦
    2.73
    Ń
    2.72
    UTERS
    2.72
    Ľ
    2.66
    ļ
    2.63
    Activations Density 0.036%

    No Known Activations