© 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
    2. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    3. GPT2-Small
    4. Residual Post - 32k
    5. 8-RES_POST_32K-OAI
    6. 4761
    Prev
    Next
    INDEX
    Explanations

    occurrences of the word "sym", particularly in contexts related to symposiums or symbiotic relationships

    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
    19
    +0.12
    0.5%
    555
    +0.10
    0.5%
    231
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    231
    +0.12
    0.01
    555
    +0.10
    0.01
    19
    +0.10
    0.01
    Negative Logits
     WARN
    -0.74
     Palmer
    -0.73
     Leilan
    -0.67
    ãģ¦
    -0.66
    FUL
    -0.65
    WARD
    -0.65
     RAW
    -0.62
     sow
    -0.62
     Kelley
    -0.62
    DERR
    -0.62
    POSITIVE LOGITS
    posium
    1.70
    phony
    1.41
    ptoms
    1.34
    pt
    1.06
    ple
    1.04
    plin
    1.01
    antics
    0.99
    pos
    0.99
    pled
    0.98
    pling
    0.96
    Activations Density 0.024%

    No Known Activations