© 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. 27862
    Prev
    Next
    INDEX
    Explanations

    references to helplines or support services for assistance

    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
    77
    +0.15
    0.9%
    387
    +0.13
    0.7%
    450
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    203
    +0.15
    0.01
    303
    +0.13
    0.01
    387
    +0.12
    0.01
    Negative Logits
    atives
    -1.82
    ables
    -1.81
     onions
    -1.55
     Socialist
    -1.55
     wool
    -1.53
    helial
    -1.53
     dignity
    -1.52
     union
    -1.49
     thereto
    -1.47
     Legislature
    -1.45
    POSITIVE LOGITS
    Ĵ
    2.18
    ppo
    1.93
    sted
    1.77
    ®
    1.69
    uso
    1.62
    bered
    1.62
    Ľ
    1.61
    ophone
    1.57
    rpm
    1.56
    strom
    1.53
    Activations Density 3.059%

    No Known Activations