© 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. 29964
    Prev
    Next
    INDEX
    Explanations

    instances of the term "chamber" 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
    156
    +0.17
    1.0%
    81
    +0.13
    0.7%
    217
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    184
    +0.17
    0.01
    0
    +0.13
    0.01
    499
    +0.12
    0.01
    Negative Logits
    ĸ´
    -2.44
    ³
    -2.41
    §
    -2.38
    ·¸
    -2.38
    ¯
    -2.37
    İ
    -2.26
    ĵ
    -2.15
    ¹
    -2.04
    Ń
    -2.03
    ½
    -2.03
    POSITIVE LOGITS
    room
    1.71
    plit
    1.67
    ingly
    1.67
    heet
    1.67
    rooms
    1.60
    naire
    1.60
    pool
    1.59
    plate
    1.59
    lot
    1.58
    tee
    1.55
    Activations Density 0.036%

    No Known Activations