© 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. 5297
    Prev
    Next
    INDEX
    Explanations

    characters or strings that represent code or programming syntax

    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
    +0.17
    0.8%
    457
    +0.10
    0.5%
    257
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    659
    +0.17
    0.04
    67
    +0.10
    0.03
    670
    +0.10
    0.02
    Negative Logits
    etheless
    -0.82
     oun
    -0.78
    ĺħ
    -0.78
     disadvant
    -0.74
    ngth
    -0.73
    icans
    -0.72
     teasp
    -0.71
     grap
    -0.71
    isin
    -0.71
    £ı
    -0.70
    POSITIVE LOGITS
    :\
    0.85
    Filename
    0.83
     src
    0.81
    psc
    0.76
    Url
    0.76
    Section
    0.75
    Domain
    0.74
    Texture
    0.72
     URL
    0.72
    src
    0.70
    Activations Density 0.164%

    No Known Activations