© 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. 24440
    Prev
    Next
    INDEX
    Explanations

    phrases that indicate changes or shifts in society or personal circumstances

    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
    193
    +0.11
    0.6%
    81
    +0.11
    0.6%
    141
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    81
    +0.11
    0.10
    193
    +0.11
    0.09
    331
    +0.10
    0.10
    Negative Logits
    ¾
    -3.15
    ĻĤ
    -3.00
    ±
    -2.94
    ¦
    -2.94
    ª
    -2.88
    Ĩ
    -2.86
    °
    -2.86
    Ĭ
    -2.84
    §
    -2.79
    ģ
    -2.77
    POSITIVE LOGITS
    forge
    1.53
    gmail
    1.47
    ento
    1.32
    arant
    1.27
    âĢ¯
    1.25
    ='
    1.24
    ="-
    1.24
    oscope
    1.22
    Scheme
    1.21
     monet
    1.21
    Activations Density 3.974%

    No Known Activations