© 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. 22799
    Prev
    Next
    INDEX
    Explanations

    phrases that express clarity or emphasis on understanding and disclosure

    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
    409
    +0.13
    0.7%
    176
    +0.12
    0.7%
    98
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    130
    +0.13
    0.09
    409
    +0.12
    0.08
    397
    +0.10
    0.08
    Negative Logits
    Īĺ
    -4.84
    ¿
    -4.58
    ¦
    -4.53
    °
    -4.39
    ¶
    -4.37
    ĻĤ
    -4.35
    ŀ
    -4.24
    ł
    -4.22
    Ĩ
    -4.16
    £
    -4.09
    POSITIVE LOGITS
    ".
    1.81
     therefrom
    1.74
    .).
    1.64
     etc
    1.55
    ressor
    1.54
    '.
    1.51
     since
    1.51
    erver
    1.50
    era
    1.49
    ":{
    1.48
    Activations Density 1.860%

    No Known Activations