© 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. 23504
    Prev
    Next
    INDEX
    Explanations

    references to licensing and legal terms associated with the Apache Software Foundation

    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
    156
    +0.26
    1.5%
    86
    +0.14
    0.8%
    294
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    479
    +0.26
    -0.01
    495
    +0.14
    0.10
    86
    +0.13
    0.03
    Negative Logits
    aterally
    -1.62
    åĬ¡
    -1.60
    ÑĢÑĥг
    -1.51
    олÑĮз
    -1.46
    mathchoice
    -1.46
    Ñĸд
    -1.44
     0000000000000000000000000000000000
    -1.40
    аÑģÑģ
    -1.39
     hereto
    -1.36
    áz
    -1.36
    POSITIVE LOGITS
     standpoint
    1.66
     Lives
    1.56
     perspective
    1.51
     chase
    1.43
     oneself
    1.43
     effect
    1.42
     term
    1.35
     tamp
    1.35
    veis
    1.32
     ni
    1.31
    Activations Density 1.860%

    No Known Activations