Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    3. Pythia-70M-Deduped
    4. MLP Post
    5. 0-MLP-SM
    6. 27608
    Prev
    Next
    INDEX
    Explanations

    instances of the underscore character

    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/0-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.0.hook_mlp_out
    Hook Layer
    0
    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
    125
    +0.12
    0.7%
    400
    +0.12
    0.6%
    296
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    354
    +0.12
    0.07
    51
    +0.12
    0.05
    103
    +0.11
    0.06
    Negative Logits
     sizes
    -1.49
     mates
    -1.46
     mate
    -1.45
    obs
    -1.40
    edly
    -1.37
     equivalents
    -1.36
     Bake
    -1.32
     yourself
    -1.30
    onna
    -1.27
     single
    -1.26
    POSITIVE LOGITS
    irection
    1.65
    CUSSION
    1.64
    ição
    1.64
    lining
    1.55
     ordinance
    1.54
    iencies
    1.54
    inian
    1.53
    orea
    1.53
     constitution
    1.51
     legislature
    1.49
    Activations Density 0.039%

    No Known Activations