© 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. 25851
    Prev
    Next
    INDEX
    Explanations

    negative sentiments or criticisms

    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
    156
    +0.36
    2.1%
    276
    +0.12
    0.7%
    223
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    276
    +0.36
    0.04
    210
    +0.12
    0.01
    141
    +0.11
    0.04
    Negative Logits
    »¿
    -3.92
    Ī
    -3.69
    į
    -3.65
    Į
    -3.55
    ·¸
    -3.49
    ĨĴ
    -3.48
    ĥ½
    -3.47
    Ļª
    -3.40
    Ĭ
    -3.32
    ¿½
    -3.32
    POSITIVE LOGITS
     trouble
    1.50
     medicine
    1.47
    wash
    1.37
     baths
    1.37
     inevitably
    1.35
    places
    1.35
     comfort
    1.33
     thread
    1.32
    woke
    1.28
     compared
    1.27
    Activations Density 0.135%

    No Known Activations