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

    Neuronpedia

    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. 26685
    Prev
    Next
    INDEX
    Explanations

    phrases indicating possession or experience of various entities

    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
    368
    +0.11
    0.6%
    305
    +0.11
    0.6%
    207
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    108
    +0.11
    0.09
    331
    +0.11
    0.08
    272
    +0.10
    0.10
    Negative Logits
     anyone
    -1.72
    ights
    -1.69
     anybody
    -1.69
    naments
    -1.66
     those
    -1.58
     thin
    -1.56
    ,''
    -1.56
     twins
    -1.54
     hers
    -1.50
     narrowing
    -1.49
    POSITIVE LOGITS
    ¼
    2.48
    £
    2.33
    ¡
    2.22
    Ĵ
    2.01
    ĺ
    2.01
    »¿
    1.96
    º
    1.94
    enez
    1.90
    Ļ
    1.90
    1.86
    Activations Density 1.686%

    No Known Activations