© 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. 25151
    Prev
    Next
    INDEX
    Explanations

    email headers

    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
    497
    +0.12
    0.6%
    313
    +0.11
    0.6%
    385
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    184
    +0.12
    0.01
    497
    +0.11
    0.00
    462
    +0.11
    0.01
    Negative Logits
    atinib
    -1.81
    jur
    -1.71
     NJ
    -1.46
    otec
    -1.42
    jo
    -1.41
    ederal
    -1.40
     prosecutor
    -1.38
     jurisdiction
    -1.37
     Racing
    -1.36
    oelectric
    -1.36
    POSITIVE LOGITS
     alike
    1.83
    âĢ¦]
    1.79
    ...]
    1.78
     liking
    1.60
    #####
    1.53
     respectively
    1.52
     successes
    1.48
     _
    1.45
    .:
    1.39
     cf
    1.37
    Activations Density 0.203%

    No Known Activations