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. 4232
    Prev
    Next
    INDEX
    Explanations

    phrases that contain the word "of" used in various contexts

    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
    393
    +0.22
    1.0%
    362
    +0.14
    0.7%
    608
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    362
    +0.22
    0.04
    692
    +0.14
    0.03
    669
    +0.12
    0.04
    Negative Logits
     recip
    -0.69
    reddits
    -0.68
     CLR
    -0.68
     inherit
    -0.64
    ghai
    -0.64
    NET
    -0.62
     Advertisement
    -0.61
    vic
    -0.60
    zai
    -0.60
    afia
    -0.60
    POSITIVE LOGITS
     nowhere
    0.89
     bounds
    0.86
     existence
    0.73
     Yugoslavia
    0.71
     orbit
    0.70
     Rowe
    0.70
    reach
    0.69
    ventory
    0.68
     sheer
    0.67
     course
    0.66
    Activations Density 0.061%

    No Known Activations