© 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. MLP Post
    9. 2-MLP-SM
    11. 24405
    Prev
    Next
    INDEX
    Explanations

    mentions of cities, particularly focusing on Chicago and Atlanta

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    ctigges/pythia-70m-deduped__mlp-sm_processed/2-mlp-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_mlp_out
    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
    23
    +0.20
    1.2%
    268
    +0.14
    0.8%
    9
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    268
    +0.20
    0.06
    9
    +0.14
    0.05
    79
    +0.13
    0.05
    Negative Logits
     ciudad
    -1.70
     forum
    -1.70
    blogger
    -1.69
     modo
    -1.59
     niche
    -1.58
     qué
    -1.55
    journals
    -1.50
    suppl
    -1.44
     way
    -1.42
     avenue
    -1.41
    POSITIVE LOGITS
    ¸
    1.75
     mercy
    1.74
    ĩ
    1.70
    Ķ
    1.69
    attery
    1.67
    axes
    1.67
    rains
    1.64
    ľ
    1.61
    ility
    1.60
    ´
    1.60
    Activations Density 0.342%

    No Known Activations