© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 26517
    Prev
    Next
    INDEX
    Explanations

    instances of the word "built" in relation to structures or constructions

    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
    168
    +0.14
    0.8%
    433
    +0.11
    0.6%
    491
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    168
    +0.14
    0.03
    433
    +0.11
    0.02
    260
    +0.10
    0.00
    Negative Logits
    Ŀ
    -1.79
    ı
    -1.74
    ±
    -1.59
    idone
    -1.55
    ubern
    -1.54
    /-/
    -1.52
    onna
    -1.49
    iate
    -1.48
     cheeks
    -1.43
    ?'
    -1.43
    POSITIVE LOGITS
     regulation
    1.41
    ante
    1.40
    cient
    1.39
     torsion
    1.37
     mistake
    1.34
     restriction
    1.33
     designation
    1.33
     \"
    1.32
    sburg
    1.32
     xe
    1.30
    Activations Density 0.279%

    No Known Activations