Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    3. GPT2-Small
    4. Residual Post - 32k
    5. 8-RES_POST_32K-OAI
    6. 5090
    Prev
    Next
    INDEX
    Explanations

    references to governmental committees

    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
    373
    +0.15
    0.7%
    88
    +0.09
    0.4%
    357
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    113
    +0.15
    0.01
    106
    +0.09
    0.01
    31
    +0.09
    0.01
    Negative Logits
     dwelling
    -0.62
     shr
    -0.61
    senal
    -0.61
     IPM
    -0.60
     pus
    -0.59
    ixel
    -0.57
     âĵĺ
    -0.57
     Cosmos
    -0.57
    ¥ŀ
    -0.56
     mosqu
    -0.56
    POSITIVE LOGITS
    BuyableInstoreAndOnline
    0.72
    upon
    0.67
    Coun
    0.67
    iem
    0.66
    Bott
    0.66
    akia
    0.65
    answer
    0.65
    Console
    0.65
     Comfort
    0.64
    Lear
    0.64
    Activations Density 0.014%

    No Known Activations