© 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 EvalsExportsAPI 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. 25074
    Prev
    Next
    INDEX
    Explanations

    elements related to events, proposals, and conditions 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/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
    452
    +0.15
    0.8%
    223
    +0.11
    0.6%
    329
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    223
    +0.15
    0.06
    452
    +0.11
    0.02
    174
    +0.10
    0.08
    Negative Logits
    ete
    -1.73
    estry
    -1.60
    ene
    -1.50
     incumbent
    -1.49
    ery
    -1.46
    elta
    -1.46
    ede
    -1.45
    esan
    -1.44
    ult
    -1.42
    zona
    -1.36
    POSITIVE LOGITS
    ²
    3.44
    ³
    3.40
    ¼
    3.39
    «
    3.31
    Ħ
    3.21
    ¿½
    3.19
    Ĥ
    3.18
    »
    3.16
    ij
    3.15
    ĸ´
    3.09
    Activations Density 2.349%

    No Known Activations