© 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. 23980
    Prev
    Next
    INDEX
    Explanations

    specific structured formats or code syntax in programming languages

    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
    221
    +0.16
    0.9%
    49
    +0.12
    0.6%
    383
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    221
    +0.16
    0.01
    474
    +0.12
    0.01
    207
    +0.10
    0.02
    Negative Logits
    ½
    -2.01
    Į
    -1.99
    º
    -1.98
    Īĺ
    -1.97
    Ĵ
    -1.89
    ¾
    -1.84
    ¹
    -1.76
    Ľ
    -1.72
    ĥ
    -1.71
    ario
    -1.70
    POSITIVE LOGITS
     ICC
    1.67
     disproportion
    1.55
     refractive
    1.51
     immunod
    1.49
     IOP
    1.48
     hypoxic
    1.44
    ively
    1.43
     Âĵ
    1.42
     capacitance
    1.42
    utely
    1.41
    Activations Density 0.166%

    No Known Activations