© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 25941
    Prev
    Next
    INDEX
    Explanations

    terms related to dynamic and static systems or concepts

    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
    420
    +0.14
    0.8%
    52
    +0.11
    0.6%
    242
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    23
    +0.14
    0.01
    37
    +0.11
    0.02
    478
    +0.10
    0.03
    Negative Logits
    ĥ½
    -2.11
    Ĥ¬
    -1.92
    ģ
    -1.78
    fest
    -1.73
    lessness
    -1.72
    Ļ
    -1.62
    İ
    -1.62
    ®
    -1.55
    ª
    -1.53
    lis
    -1.52
    POSITIVE LOGITS
     clear
    1.57
     doubt
    1.48
    ient
    1.47
     Woman
    1.39
     men
    1.37
     categor
    1.35
     constitution
    1.32
     second
    1.30
     my
    1.29
     woman
    1.29
    Activations Density 0.252%

    No Known Activations