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. 4257
    Prev
    Next
    INDEX
    Explanations

    periods and punctuation marks

    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
    373
    +0.19
    0.8%
    314
    +0.10
    0.4%
    110
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    107
    +0.19
    0.02
    308
    +0.10
    0.01
    456
    +0.09
    0.01
    Negative Logits
    ãĥ´
    -0.80
     Pixie
    -0.65
    éĹĺ
    -0.62
     insol
    -0.61
    ypes
    -0.60
     fortun
    -0.59
     muff
    -0.57
    wcs
    -0.57
     surrounds
    -0.56
     impart
    -0.56
    POSITIVE LOGITS
    actionDate
    0.83
    Roberts
    0.77
    mil
    0.76
    kat
    0.70
    rium
    0.68
    duction
    0.68
    tis
    0.64
    Vote
    0.64
    GI
    0.63
    FTWARE
    0.62
    Activations Density 0.101%

    No Known Activations