© 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
    3. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    5. GPT2-Small
    7. Residual Post - 32k
    9. 8-RES_POST_32K-OAI
    11. 5001
    Prev
    Next
    INDEX
    Explanations

    instances of loudness or descriptions related to noise levels

    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
    747
    +0.09
    0.4%
    12
    +0.08
    0.4%
    330
    +0.08
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    271
    +0.09
    0.02
    553
    +0.08
    0.01
    70
    +0.08
    0.02
    Negative Logits
    onz
    -0.76
     Conditions
    -0.66
     Framework
    -0.63
     muster
    -0.63
     Korra
    -0.63
     Gould
    -0.63
    CG
    -0.61
     Sutherland
    -0.60
     Daredevil
    -0.59
    ANA
    -0.58
    POSITIVE LOGITS
    mouth
    1.34
    est
    1.18
    ness
    0.98
    eners
    0.97
    ifying
    0.84
    ening
    0.83
    ened
    0.83
    ener
    0.82
     Enough
    0.79
    geoning
    0.79
    Activations Density 0.099%

    No Known Activations