Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    3. Pythia-70M-Deduped
    4. Residuals
    5. 1-RES-SM
    6. 26833
    Prev
    Next
    INDEX
    Explanations

    terms related to ultra-high performance or cutting-edge technology

    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/1-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.1.hook_resid_post
    Hook Layer
    1
    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
    203
    +0.16
    0.9%
    111
    +0.16
    0.9%
    502
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    111
    +0.16
    0.07
    271
    +0.16
    0.03
    156
    +0.13
    0.04
    Negative Logits
     principal
    -1.75
     mutual
    -1.62
     false
    -1.61
     strict
    -1.58
     frequent
    -1.58
     particular
    -1.50
     discipline
    -1.48
     forget
    -1.43
     wont
    -1.43
     pupils
    -1.43
    POSITIVE LOGITS
    ĻĤ
    4.05
    ī
    3.91
    ¦
    3.60
    ·
    3.56
    Ĵ
    3.51
    ¤
    3.50
    ģ
    3.46
    Ĩ
    3.41
    IJ
    3.38
    ħ
    3.35
    Activations Density 0.165%

    No Known Activations