© 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. 4993
    Prev
    Next
    INDEX
    Explanations

    keywords and phrases related to scientific theories and historical scientific figures

    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
    760
    +0.13
    0.6%
    420
    +0.10
    0.5%
    388
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    284
    +0.13
    0.01
    314
    +0.10
    -0.00
    17
    +0.10
    0.01
    Negative Logits
    ĸļ
    -0.84
    ĪĴ
    -0.82
    ãĥ¯ãĥ³
    -0.76
    Offline
    -0.75
    rals
    -0.70
    phies
    -0.69
    ById
    -0.68
    Joined
    -0.68
    adelphia
    -0.66
    legates
    -0.66
    POSITIVE LOGITS
    potion
    1.06
    demon
    0.83
     fung
    0.79
    ffect
    0.78
     mutation
    0.77
     immortality
    0.76
     syndrome
    0.76
     longevity
    0.75
     fertility
    0.69
     beneficial
    0.69
    Activations Density 0.034%

    No Known Activations