© 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. 4985
    Prev
    Next
    INDEX
    Explanations

    references to document identification or publication numbering

    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
    +0.16
    0.7%
    497
    +0.11
    0.5%
    158
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    754
    +0.16
    0.02
    616
    +0.11
    0.02
    311
    +0.11
    0.01
    Negative Logits
    aimon
    -0.70
     outp
    -0.61
     glances
    -0.61
     halftime
    -0.61
    iem
    -0.59
     proxies
    -0.59
    itu
    -0.59
     Glow
    -0.58
    tein
    -0.58
    ricular
    -0.57
    POSITIVE LOGITS
    XXXX
    0.65
    ortment
    0.64
    manufact
    0.64
     228
    0.63
     259
    0.63
    Manufact
    0.62
     258
    0.62
    217
    0.61
     655
    0.61
     Marshal
    0.59
    Activations Density 0.072%

    No Known Activations