© 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. 23472
    Prev
    Next
    INDEX
    Explanations

    terms related to electrical charge and energy management

    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
    345
    +0.13
    0.7%
    365
    +0.12
    0.6%
    468
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    467
    +0.13
    0.02
    345
    +0.12
    0.03
    100
    +0.11
    0.04
    Negative Logits
    stones
    -1.66
    ItemGroup
    -1.59
    nai
    -1.51
    keywords
    -1.50
    stone
    -1.29
    :.
    -1.25
     topic
    -1.25
    TypeDef
    -1.24
    hill
    -1.22
    zzo
    -1.22
    POSITIVE LOGITS
    ](
    1.58
    hift
    1.45
     least
    1.42
    ")]
    1.41
    otted
    1.41
    isan
    1.40
    arak
    1.40
    afe
    1.39
    antic
    1.39
    ample
    1.33
    Activations Density 0.403%

    No Known Activations