© 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. 25136
    Prev
    Next
    INDEX
    Explanations

    prepositions and their occurrences in sentences

    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
    66
    +0.12
    0.7%
    9
    +0.10
    0.6%
    151
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    133
    +0.12
    0.03
    151
    +0.10
    0.03
    66
    +0.10
    0.03
    Negative Logits
    icy
    -1.68
     among
    -1.59
     descent
    -1.59
    dit
    -1.41
    chin
    -1.39
     clot
    -1.37
     duo
    -1.34
     carcinogenesis
    -1.32
     Parenthood
    -1.31
     trier
    -1.30
    POSITIVE LOGITS
    ures
    1.76
    ables
    1.65
    URES
    1.56
    fered
    1.52
    STS
    1.47
    ĥ½
    1.42
    bies
    1.36
    ways
    1.35
    GTH
    1.34
     Rey
    1.34
    Activations Density 0.159%

    No Known Activations