© 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. 4873
    Prev
    Next
    INDEX
    Explanations

    avocations related to names or identifiers associated with individuals or entities

    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
    314
    +0.16
    0.7%
    27
    +0.13
    0.6%
    739
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    314
    +0.16
    0.01
    23
    +0.13
    0.06
    107
    +0.11
    0.05
    Negative Logits
     whiff
    -0.82
     Ples
    -0.69
     scar
    -0.66
     Canary
    -0.66
     cig
    -0.64
     yeast
    -0.61
     overc
    -0.61
     lob
    -0.61
     phr
    -0.61
     Monstrous
    -0.61
    POSITIVE LOGITS
    Depth
    0.89
    Mine
    0.89
    arthed
    0.88
    itors
    0.86
    ition
    0.86
    metics
    0.85
    Dem
    0.85
    Aren
    0.84
    Energy
    0.83
    Rate
    0.83
    Activations Density 0.707%

    No Known Activations