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

    Neuronpedia

    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. 25527
    Prev
    Next
    INDEX
    Explanations

    characteristics or descriptors that denote significance, excellence, or uniqueness 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/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
    272
    +0.18
    1.0%
    33
    +0.11
    0.6%
    179
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    179
    +0.18
    0.09
    272
    +0.11
    0.10
    248
    +0.11
    0.06
    Negative Logits
     pounds
    -1.85
    iners
    -1.73
    verse
    -1.59
    pect
    -1.48
    nej
    -1.42
    itar
    -1.39
    osities
    -1.38
     (_
    -1.38
     Treaty
    -1.38
    ns
    -1.33
    POSITIVE LOGITS
    ĻĤ
    2.86
    Ī
    2.42
    Īĺ
    2.31
    ľĵ
    2.20
    ¡
    2.09
    ¿
    2.02
    ĨĴ
    2.00
    Ľ
    1.99
    »¿
    1.95
    ĭ
    1.94
    Activations Density 1.051%

    No Known Activations