© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. Residuals
    9. 1-RES-SM
    11. 28324
    Prev
    Next
    INDEX
    Explanations

    terms related to railways and train transportation

    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
    156
    +0.19
    1.1%
    375
    +0.13
    0.7%
    114
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    404
    +0.19
    0.02
    395
    +0.13
    0.02
    45
    +0.12
    0.02
    Negative Logits
     pray
    -1.50
    nesty
    -1.48
     hope
    -1.46
     apologize
    -1.43
     cytokines
    -1.40
     seed
    -1.38
     fasting
    -1.37
    entional
    -1.35
    7554
    -1.35
     sunlight
    -1.35
    POSITIVE LOGITS
    bed
    1.69
    ŀ
    1.68
    Station
    1.68
    craft
    1.67
     Station
    1.60
     entry
    1.59
     infrastructure
    1.58
    roads
    1.57
    hors
    1.53
    gebra
    1.52
    Activations Density 0.150%

    No Known Activations