© 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
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. MLP Post
    9. 2-MLP-SM
    11. 25987
    Prev
    Next
    INDEX
    Explanations

    negative aspects or outcomes associated with legal cases

    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
    218
    +0.13
    0.7%
    23
    +0.12
    0.6%
    167
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    218
    +0.13
    0.02
    97
    +0.12
    0.02
    511
    +0.11
    0.01
    Negative Logits
     "</
    -1.53
    stairs
    -1.52
    findViewById
    -1.48
     caption
    -1.46
    áĢ¯
    -1.32
    ив
    -1.29
    biamo
    -1.29
     vid
    -1.29
    áĢ¬
    -1.27
     secre
    -1.27
    POSITIVE LOGITS
    ake
    1.89
    akes
    1.80
    ries
    1.76
    arers
    1.70
     oneself
    1.62
    ality
    1.62
    iors
    1.57
    inners
    1.50
    ites
    1.47
    ring
    1.44
    Activations Density 0.055%

    No Known Activations