© 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. 25619
    Prev
    Next
    INDEX
    Explanations

    formal or structured references to court cases or legal documents

    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
    391
    +0.15
    0.9%
    165
    +0.13
    0.8%
    400
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    400
    +0.15
    0.05
    486
    +0.13
    0.03
    165
    +0.13
    0.05
    Negative Logits
    »¿
    -1.77
    heet
    -1.65
    bum
    -1.56
    heed
    -1.40
    ids
    -1.40
    Stra
    -1.39
    stra
    -1.38
    had
    -1.37
    itic
    -1.36
    olar
    -1.36
    POSITIVE LOGITS
    ncia
    1.58
     moderator
    1.52
     [âĢ¦]
    1.49
     ger
    1.48
     “
    1.40
     Lev
    1.33
     parcel
    1.32
    &#
    1.32
     aged
    1.31
     Surviv
    1.31
    Activations Density 0.174%

    No Known Activations