© 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 EvalsExportsAPI 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. 24917
    Prev
    Next
    INDEX
    Explanations

    mentions of the word "first" 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
    156
    +0.13
    0.7%
    476
    +0.11
    0.6%
    92
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    15
    +0.13
    0.05
    248
    +0.11
    0.04
    506
    +0.11
    0.04
    Negative Logits
     progress
    -1.53
    sets
    -1.49
    TRODUCTION
    -1.44
    umab
    -1.44
     contempt
    -1.39
     bac
    -1.36
     contradict
    -1.34
     trouble
    -1.34
     passages
    -1.34
     comprehension
    -1.34
    POSITIVE LOGITS
    ¨
    3.98
    º
    3.92
    ®
    3.87
    ģ
    3.69
    ¶
    3.63
    ¿
    3.61
    ĥ
    3.60
    ¯
    3.58
    ¬
    3.52
    ħ
    3.49
    Activations Density 0.195%

    No Known Activations