Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    3. GPT2-Small
    4. Residual Post - 32k
    5. 6-RES_POST_32K-OAI
    6. 4239
    Prev
    Next
    INDEX
    Explanations

    numerical data, especially related to statistics or measurements

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 6-res_post_32k-oai
    Configuration
    jbloom/GPT2-Small-OAI-v5-32k-resid-post-SAEs/v5_32k_layer_6.pt
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.6.hook_resid_post
    Hook Layer
    6
    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
    717
    +0.13
    0.6%
    504
    +0.11
    0.5%
    59
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    462
    +0.13
    0.06
    176
    +0.11
    0.04
    325
    +0.09
    0.06
    Negative Logits
    ĸļ
    -0.86
     omission
    -0.70
    ysis
    -0.63
     constitu
    -0.62
    sbm
    -0.61
    soDeliveryDate
    -0.61
    isance
    -0.60
     inconsistencies
    -0.60
    LOAD
    -0.59
    imaru
    -0.59
    POSITIVE LOGITS
    ruary
    0.76
    icipated
    0.73
    loe
    0.72
     Entered
    0.71
    gall
    0.70
    sold
    0.67
    luster
    0.67
    illet
    0.67
    agall
    0.66
    urdue
    0.66
    Activations Density 0.830%

    No Known Activations