© 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. 27857
    Prev
    Next
    INDEX
    Explanations

    instances of the word "bent" in various contexts

    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
    +0.14
    0.8%
    389
    +0.12
    0.6%
    148
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    333
    +0.14
    0.01
    476
    +0.12
    0.01
    317
    +0.12
    0.01
    Negative Logits
     Sydney
    -1.58
     slightest
    -1.55
     happening
    -1.51
     straight
    -1.46
    CTOR
    -1.44
     Night
    -1.42
     appearances
    -1.42
     Nile
    -1.42
     Mumbai
    -1.40
     Sah
    -1.40
    POSITIVE LOGITS
    ocument
    1.71
    ressional
    1.68
    upgreek
    1.65
    estinal
    1.62
    uant
    1.61
    probe
    1.61
    ovo
    1.61
    warf
    1.58
    endent
    1.57
    irection
    1.57
    Activations Density 0.022%

    No Known Activations