© 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. 0-MLP-SM
    6. 27157
    Prev
    Next
    INDEX
    Explanations

    instances of the word "Ros" in various contexts

    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/0-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.0.hook_mlp_out
    Hook Layer
    0
    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
    369
    +0.21
    1.2%
    376
    +0.13
    0.7%
    410
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    437
    +0.21
    0.01
    429
    +0.13
    0.01
    285
    +0.13
    0.01
    Negative Logits
    Ļª
    -1.75
     younger
    -1.69
     older
    -1.65
    heres
    -1.60
    illance
    -1.53
    »¿
    -1.50
    ĥ
    -1.48
     susp
    -1.45
    ria
    -1.45
     adult
    -1.42
    POSITIVE LOGITS
    genstein
    1.94
    coe
    1.75
    etta
    1.75
    enthal
    1.74
    away
    1.73
    Ä«
    1.54
    ifer
    1.52
    ue
    1.46
     numbered
    1.42
    udi
    1.40
    Activations Density 0.072%

    No Known Activations