© 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
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. MLP Post
    9. 2-MLP-SM
    11. 25924
    Prev
    Next
    INDEX
    Explanations

    keywords that indicate critical factors or essential elements 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
    500
    +0.13
    0.7%
    327
    +0.12
    0.6%
    118
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    500
    +0.13
    0.02
    193
    +0.12
    0.02
    118
    +0.12
    0.03
    Negative Logits
    erd
    -1.81
    iels
    -1.68
    apine
    -1.66
    rfloor
    -1.59
    yyyy
    -1.47
    ÅĦ
    -1.46
    infty
    -1.41
    snap
    -1.40
     Dogs
    -1.34
     snaps
    -1.33
    POSITIVE LOGITS
    rapy
    1.52
    ractor
    1.44
     contributor
    1.39
    heets
    1.38
     Medicare
    1.33
     tracts
    1.33
    tract
    1.31
     macro
    1.30
     accounts
    1.29
     lately
    1.29
    Activations Density 0.296%

    No Known Activations