© 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. 2-MLP-SM
    11. 24991
    Prev
    Next
    INDEX
    Explanations

    inquiries and notes related to conditions or requirements

    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
    117
    +0.12
    0.7%
    397
    +0.11
    0.6%
    222
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    397
    +0.12
    0.09
    335
    +0.11
    0.09
    326
    +0.11
    0.02
    Negative Logits
     suffice
    -1.59
     plenty
    -1.59
     likewise
    -1.58
     tolerated
    -1.51
     exactly
    -1.49
     always
    -1.49
    ertainly
    -1.48
     Parenthood
    -1.47
     randomization
    -1.39
     singleton
    -1.36
    POSITIVE LOGITS
    áĢº
    1.65
    áĢ¬
    1.51
    ulation
    1.44
    feit
    1.42
    conscious
    1.42
    space
    1.39
    iller
    1.37
    isión
    1.37
    voice
    1.37
    scene
    1.36
    Activations Density 1.698%

    No Known Activations