© 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. 24367
    Prev
    Next
    INDEX
    Explanations

    references to GitHub and the associated testing framework

    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
    199
    +0.11
    0.7%
    399
    +0.11
    0.7%
    19
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    56
    +0.11
    0.03
    163
    +0.11
    0.01
    199
    +0.11
    0.01
    Negative Logits
     ruler
    -1.45
     predecessor
    -1.44
     royalty
    -1.43
     socks
    -1.38
     cookies
    -1.37
     security
    -1.37
    cember
    -1.34
     beginnings
    -1.34
     arches
    -1.33
     debut
    -1.32
    POSITIVE LOGITS
    Ľ
    2.41
    ģ
    2.40
    ĸ
    2.38
    Ķ
    2.23
    Ł
    2.23
    ¿½
    2.23
    ĵ
    2.22
    ±
    2.21
    ľĵ
    2.17
    Ĥ
    2.16
    Activations Density 0.317%

    No Known Activations