© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    3. Pythia-70M-Deduped
    4. MLP Post
    5. 2-MLP-SM
    6. 26007
    Prev
    Next
    INDEX
    Explanations

    terms related to negative effects or harmful impacts

    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
    371
    +0.13
    0.7%
    156
    +0.11
    0.6%
    342
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    371
    +0.13
    0.02
    122
    +0.11
    0.01
    446
    +0.11
    0.01
    Negative Logits
    formed
    -1.59
    omy
    -1.54
    ugu
    -1.52
     killed
    -1.49
    ice
    -1.48
    union
    -1.47
    ploid
    -1.43
    ellee
    -1.43
    ipart
    -1.42
    ceptions
    -1.41
    POSITIVE LOGITS
     contrast
    1.50
    ŀ
    1.49
    ©
    1.47
     example
    1.45
     appraisal
    1.42
     Contents
    1.42
     contrasts
    1.41
     amendments
    1.39
     redist
    1.39
     architectural
    1.38
    Activations Density 0.116%

    No Known Activations