© 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. 0-MLP-SM
    11. 28957
    Prev
    Next
    INDEX
    Explanations

    the word "hereby."

    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/0-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.0.hook_mlp_out
    Hook Layer
    0
    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
    305
    +0.14
    0.8%
    403
    +0.14
    0.8%
    78
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    78
    +0.14
    0.01
    305
    +0.14
    0.02
    403
    +0.13
    0.02
    Negative Logits
    ´
    -2.72
    ·
    -2.52
    º
    -2.43
    ĥ½
    -2.40
    ®
    -2.29
    ¯
    -2.20
    ħ
    -2.19
    ³
    -2.17
    ĭ
    -2.11
    ij
    -2.07
    POSITIVE LOGITS
     sure
    1.70
    veis
    1.64
     condemn
    1.49
     appeals
    1.48
     recognise
    1.48
     agg
    1.46
     crop
    1.43
     heirs
    1.42
    pril
    1.39
     justify
    1.38
    Activations Density 0.138%

    No Known Activations