© 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. 22791
    Prev
    Next
    INDEX
    Explanations

    phrases indicating time, particularly focusing on references to 'the end of' in various contexts

    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
    246
    +0.12
    0.7%
    157
    +0.12
    0.7%
    300
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    300
    +0.12
    0.03
    246
    +0.12
    0.05
    492
    +0.11
    0.04
    Negative Logits
    eman
    -1.68
    ver
    -1.53
     opposite
    -1.39
     exits
    -1.37
     moment
    -1.33
    eling
    -1.33
     preced
    -1.32
    illance
    -1.32
    ild
    -1.30
    oms
    -1.30
    POSITIVE LOGITS
    claimed
    1.78
    ('#
    1.49
    claim
    1.49
    ships
    1.43
    fileID
    1.41
    assertEqual
    1.40
    testing
    1.40
    blr
    1.39
    oubtedly
    1.39
    taient
    1.37
    Activations Density 0.378%

    No Known Activations