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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    3. Pythia-70M-Deduped
    4. MLP Post
    5. 0-MLP-SM
    6. 26904
    Prev
    Next
    INDEX
    Explanations

    a specific phrase or term that signifies a part of the text related to inclusivity or belonging

    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
    376
    +0.15
    0.9%
    410
    +0.15
    0.8%
    5
    +0.14
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    5
    +0.15
    0.01
    270
    +0.15
    0.01
    70
    +0.14
    0.01
    Negative Logits
    Ĥ
    -1.75
    Ĺ
    -1.72
    ¼
    -1.61
    Ľ
    -1.49
     inches
    -1.46
    ľ
    -1.45
    <|outofrange|>
    -1.38
    ↵             
    -1.38
    ↵                               
    -1.38
    <|outofrange|>
    -1.38
    POSITIVE LOGITS
    leneck
    1.70
    lee
    1.60
    ilers
    1.45
    onian
    1.44
    blood
    1.43
    ENTIAL
    1.42
    TERN
    1.42
    eur
    1.41
    ête
    1.41
    als
    1.40
    Activations Density 0.265%

    No Known Activations