© 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. 28051
    Prev
    Next
    INDEX
    Explanations

    the word "ile" 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
    70
    +0.13
    0.7%
    453
    +0.12
    0.7%
    410
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    488
    +0.13
    0.01
    203
    +0.12
    0.01
    453
    +0.12
    0.01
    Negative Logits
     less
    -1.78
    Figure
    -1.55
    noreply
    -1.55
    ?_
    -1.52
     fewer
    -1.51
     respectively
    -1.42
    FIG
    -1.40
     between
    -1.39
    itter
    -1.38
    nez
    -1.37
    POSITIVE LOGITS
    ĻĤ
    3.31
    Ģ
    2.61
    ®
    2.40
    Ĩ
    2.33
    ħ
    2.24
    «
    2.15
    ©
    2.10
    ¬
    2.10
    º
    2.08
    ½
    2.08
    Activations Density 3.094%

    No Known Activations