© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 4455
    Prev
    Next
    INDEX
    Explanations

    mentions of the word "off" in various contexts

    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
    +0.14
    0.6%
    61
    +0.10
    0.5%
    373
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    696
    +0.14
    0.03
    533
    +0.10
    0.03
    214
    +0.10
    0.02
    Negative Logits
    Untitled
    -0.65
     Brach
    -0.61
     Pione
    -0.61
     Defender
    -0.59
     duct
    -0.57
    killer
    -0.57
     Beetle
    -0.55
     Rapt
    -0.55
     dread
    -0.55
     Theft
    -0.54
    POSITIVE LOGITS
    ices
    0.97
    loaded
    0.90
    beat
    0.85
    hand
    0.84
    load
    0.82
    loading
    0.81
    limits
    0.80
    ICE
    0.79
    cuts
    0.79
    lane
    0.78
    Activations Density 0.058%

    No Known Activations