© 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. 25984
    Prev
    Next
    INDEX
    Explanations

    technical descriptions of structural and mechanical components

    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
    49
    +0.14
    0.8%
    431
    +0.11
    0.6%
    163
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    38
    +0.14
    0.01
    347
    +0.11
    0.03
    163
    +0.11
    0.01
    Negative Logits
     hereinafter
    -1.77
     <-
    -1.75
     rendition
    -1.58
     reproduction
    -1.58
     success
    -1.57
     (),
    -1.51
     species
    -1.46
     invention
    -1.46
     compensation
    -1.45
    ially
    -1.40
    POSITIVE LOGITS
    amazon
    1.53
    enches
    1.52
    iten
    1.52
    cdn
    1.48
    aways
    1.46
    kered
    1.46
     lanes
    1.43
     teaspoons
    1.33
    enÃŃ
    1.30
    en
    1.30
    Activations Density 0.202%

    No Known Activations