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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. MLP Post
    9. 2-MLP-SM
    11. 24028
    Prev
    Next
    INDEX
    Explanations

    instances of numeric values or symbols that denote data

    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/2-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.2.hook_mlp_out
    Hook Layer
    2
    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
    +0.28
    1.6%
    23
    +0.12
    0.7%
    462
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    462
    +0.28
    0.01
    23
    +0.12
    0.01
    424
    +0.12
    0.01
    Negative Logits
    ³
    -3.52
    »¿
    -3.50
    ī
    -3.46
    Ĭ
    -3.45
    ľĵ
    -3.40
    Ħ
    -3.19
    ĩ
    -3.18
    ↵
    -3.07
    ↵  âĢĥ
    -3.07
    -3.07
    POSITIVE LOGITS
     indeed
    1.45
    named
    1.39
     crash
    1.37
     FL
    1.37
     hell
    1.35
     paradise
    1.33
     AFL
    1.32
    ].)
    1.31
     blast
    1.31
    ters
    1.29
    Activations Density 0.093%

    No Known Activations