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. 27503
    Prev
    Next
    INDEX
    Explanations

    specialized codes or references in programming contexts, particularly related to libraries or packages

    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
    410
    +0.21
    1.2%
    369
    +0.20
    1.2%
    203
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    410
    +0.21
    0.02
    111
    +0.20
    0.02
    214
    +0.12
    0.01
    Negative Logits
    pecially
    -1.70
    atin
    -1.65
    ighting
    -1.62
    ights
    -1.62
    illance
    -1.59
    ighter
    -1.56
    ibility
    -1.51
    piring
    -1.51
    porter
    -1.49
    oder
    -1.49
    POSITIVE LOGITS
    ĻĤ
    1.73
    nan
    1.50
    imm
    1.46
    meg
    1.45
    ops
    1.35
    trees
    1.32
     pointers
    1.31
    SHA
    1.30
     bears
    1.30
    bugs
    1.30
    Activations Density 0.020%

    No Known Activations