© 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. 23821
    Prev
    Next
    INDEX
    Explanations

    instances or comments in code where ellipses are used

    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
    435
    +0.12
    0.7%
    431
    +0.12
    0.6%
    151
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    385
    +0.12
    0.03
    163
    +0.12
    0.01
    24
    +0.11
    0.02
    Negative Logits
    ħ
    -2.41
    ²
    -2.35
    ¨
    -2.26
    °
    -2.24
    ĥ½
    -2.22
    ¼
    -2.19
    ı
    -2.18
    -2.14
    ↵↵↵
    -2.14
    -2.14
    POSITIVE LOGITS
    mouth
    1.49
    ace
    1.44
    asta
    1.43
    este
    1.40
     which
    1.38
    Ro
    1.36
    arium
    1.35
    strom
    1.33
     whom
    1.32
     fres
    1.31
    Activations Density 0.137%

    No Known Activations