© 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. 23148
    Prev
    Next
    INDEX
    Explanations

    punctuation marks, indicative of sentence endings

    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
    370
    +0.13
    0.7%
    218
    +0.12
    0.7%
    6
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    370
    +0.13
    0.05
    424
    +0.12
    0.03
    486
    +0.11
    0.02
    Negative Logits
    »¿
    -1.76
    oler
    -1.69
    bage
    -1.68
    ĻĤ
    -1.68
     hereby
    -1.57
    icago
    -1.56
     thereof
    -1.51
    ands
    -1.49
     proxim
    -1.45
    enron
    -1.40
    POSITIVE LOGITS
    WN
    1.69
    ping
    1.63
    Äĩ
    1.58
    xit
    1.57
    ties
    1.52
    eu
    1.50
    uration
    1.48
     Surviv
    1.43
    ÅĻ
    1.42
    mark
    1.40
    Activations Density 0.146%

    No Known Activations