© 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. 2-MLP-SM
    11. 22825
    Prev
    Next
    INDEX
    Explanations

    terms related to limitations and restrictions

    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
    469
    +0.16
    0.9%
    387
    +0.15
    0.8%
    369
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    454
    +0.16
    0.03
    440
    +0.15
    0.02
    404
    +0.12
    0.10
    Negative Logits
     adopted
    -1.66
     promul
    -1.62
    isman
    -1.55
     Amend
    -1.52
    itness
    -1.45
    olite
    -1.43
     adopting
    -1.43
    velopment
    -1.39
     reliance
    -1.37
    rios
    -1.37
    POSITIVE LOGITS
    ĻĤ
    1.71
    }$\
    1.53
    processing
    1.45
    illion
    1.40
    tons
    1.38
    tid
    1.35
    together
    1.34
    ľĵ
    1.32
     birthday
    1.31
    ($
    1.31
    Activations Density 3.003%

    No Known Activations