© 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
    2. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    3. Pythia-70M-Deduped
    4. MLP Post
    5. 2-MLP-SM
    6. 26012
    Prev
    Next
    INDEX
    Explanations

    specific terms related to time periods and execution status in programming 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
    349
    +0.11
    0.6%
    244
    +0.11
    0.6%
    278
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    23
    +0.11
    0.04
    56
    +0.11
    0.09
    374
    +0.10
    0.01
    Negative Logits
     representative
    -1.58
    strom
    -1.55
    "}](#
    -1.53
     expressed
    -1.53
    iator
    -1.48
     outside
    -1.44
     disclosed
    -1.40
     outlet
    -1.40
     quarters
    -1.39
    '):
    -1.39
    POSITIVE LOGITS
    ope
    1.81
    opes
    1.65
    cludes
    1.47
     penicillin
    1.44
    TestCase
    1.39
    angs
    1.37
    smallcaps
    1.37
    ât
    1.35
    ureus
    1.35
     doubt
    1.34
    Activations Density 1.511%

    No Known Activations