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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. MLP Post
    9. 0-MLP-SM
    11. 26730
    Prev
    Next
    INDEX
    Explanations

    occurrences of the word "join."

    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
    376
    +0.18
    1.0%
    463
    +0.12
    0.7%
    115
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    401
    +0.18
    0.02
    124
    +0.12
    0.02
    347
    +0.11
    0.02
    Negative Logits
     particularly
    -1.64
     brow
    -1.59
    hire
    -1.51
    Microsoft
    -1.48
    OWN
    -1.42
    Competing
    -1.33
    woke
    -1.32
     celebrated
    -1.32
    COP
    -1.30
    Mail
    -1.29
    POSITIVE LOGITS
    ¼
    3.55
    °
    3.38
    Ħ
    3.24
    ¾
    3.24
    ¡
    3.24
    ī
    3.22
    µ
    3.22
    Į
    3.17
    ↵
    3.12
    <|outofrange|>
    3.12
    Activations Density 0.069%

    No Known Activations