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. 4248
    Prev
    Next
    INDEX
    Explanations

    references to financial support or funding

    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
    269
    +0.11
    0.5%
    218
    +0.11
    0.5%
    288
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    352
    +0.11
    0.03
    306
    +0.11
    0.03
    225
    +0.09
    0.03
    Negative Logits
    selves
    -0.83
     Halls
    -0.73
    ique
    -0.72
    ilus
    -0.71
    ierre
    -0.67
    chant
    -0.66
    iliary
    -0.64
    imer
    -0.64
     Ple
    -0.63
    terior
    -0.63
    POSITIVE LOGITS
    ament
    0.86
     financed
    0.79
     transfers
    0.77
    funded
    0.77
     sources
    0.77
    acion
    0.76
    amental
    0.76
     Funding
    0.75
     laundering
    0.74
     allocations
    0.74
    Activations Density 0.079%

    No Known Activations