© 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. 5306
    Prev
    Next
    INDEX
    Explanations

    references to specific professional sports positions and associated player names

    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
    175
    +0.11
    0.5%
    383
    +0.10
    0.5%
    313
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    659
    +0.11
    0.07
    314
    +0.10
    -0.01
    232
    +0.09
    0.08
    Negative Logits
    lished
    -0.70
    phis
    -0.67
    @#&
    -0.67
     |--
    -0.64
     appropriated
    -0.60
    rawdownloadcloneembedreportprint
    -0.60
    cffff
    -0.60
    ful
    -0.59
    esy
    -0.58
    DEF
    -0.57
    POSITIVE LOGITS
    oil
    0.78
    anes
    0.78
    aque
    0.77
    iner
    0.73
    antly
    0.72
    ots
    0.68
    inas
    0.68
    ined
    0.67
    iable
    0.67
    ocity
    0.67
    Activations Density 1.493%

    No Known Activations