© 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. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    5. GPT2-Small
    7. Residual Post - 32k
    9. 8-RES_POST_32K-OAI
    11. 5502
    Prev
    Next
    INDEX
    Explanations

    references to numeric identifiers and data structures related to programming

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_post_32k-oai
    Configuration
    jbloom/GPT2-Small-OAI-v5-32k-resid-post-SAEs/v5_32k_layer_8.pt
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.8.hook_resid_post
    Hook Layer
    8
    Architecture
    standard
    Context Size
    64
    Dataset
    Skylion007/openwebtext
    Activation Function
    topk
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    218
    +0.13
    0.6%
    314
    +0.11
    0.5%
    403
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    6
    +0.13
    0.16
    107
    +0.11
    0.11
    314
    +0.11
    -0.02
    Negative Logits
     mathemat
    -0.72
     paralle
    -0.70
    clusively
    -0.69
     reverber
    -0.69
     soluble
    -0.69
     negatives
    -0.66
     gran
    -0.64
     stabilization
    -0.63
     crochet
    -0.63
     damp
    -0.63
    POSITIVE LOGITS
    who
    1.28
     whom
    1.26
    ername
    1.03
     who
    0.95
     Username
    0.91
    whose
    0.90
     Cosponsors
    0.87
    gets
    0.86
    :{
    0.84
    opausal
    0.81
    Activations Density 3.099%

    No Known Activations