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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsBlogSlackPrivacy & 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. 27461
    Prev
    Next
    INDEX
    Explanations

    instances of the word "official."

    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
    156
    +0.20
    1.2%
    362
    +0.14
    0.8%
    87
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    446
    +0.20
    0.03
    208
    +0.14
    0.03
    266
    +0.12
    0.02
    Negative Logits
     latter
    -1.55
    obacterium
    -1.53
    oyle
    -1.45
    rose
    -1.44
    angers
    -1.39
    abeth
    -1.38
    anson
    -1.38
    ellow
    -1.34
    gado
    -1.34
    yman
    -1.33
    POSITIVE LOGITS
    dom
    1.98
    doms
    1.76
    ships
    1.73
    pieces
    1.67
     bodies
    1.56
    blems
    1.52
    endar
    1.50
    hood
    1.49
    ities
    1.46
    esses
    1.44
    Activations Density 0.177%

    No Known Activations