© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    3. GPT2-Small
    4. Residual Post - 32k
    5. 10-RES_POST_32K-OAI
    6. 29531
    Prev
    Next
    INDEX
    Explanations

    patterns related to specific numerical or coded references

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res_post_32k-oai
    Configuration
    jbloom/GPT2-Small-OAI-v5-32k-resid-post-SAEs/v5_32k_layer_10.pt
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.10.hook_resid_post
    Hook Layer
    10
    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
    +0.20
    0.9%
    731
    +0.11
    0.5%
    648
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    373
    +0.20
    -0.02
    659
    +0.11
    0.03
    631
    +0.10
    0.02
    Negative Logits
     ACTIONS
    -0.70
    urat
    -0.65
     Participant
    -0.62
    auga
    -0.62
    onen
    -0.58
    alos
    -0.58
    othy
    -0.56
    ulhu
    -0.56
    oine
    -0.56
     Cosponsors
    -0.55
    POSITIVE LOGITS
    ãĤ·ãĥ£
    0.64
    cler
    0.60
    age
    0.58
    ר
    0.58
     Celsius
    0.57
    TV
    0.56
    change
    0.55
    xp
    0.53
    ¯¯¯¯
    0.53
    Build
    0.53
    Activations Density 0.123%

    No Known Activations