Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    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. 5108
    Prev
    Next
    INDEX
    Explanations

    occurrences of the word "first."

    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
    244
    +0.10
    0.5%
    514
    +0.09
    0.4%
    545
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    328
    +0.10
    0.02
    545
    +0.09
    0.02
    532
    +0.09
    0.02
    Negative Logits
    WER
    -0.73
    arcity
    -0.64
    asus
    -0.63
    Tags
    -0.62
    bys
    -0.60
    \\\\\\\\
    -0.60
     perpetual
    -0.59
     Kinnikuman
    -0.58
     Yesterday
    -0.58
    Ry
    -0.58
    POSITIVE LOGITS
     embark
    0.73
    imaru
    0.66
     imm
    0.66
     glance
    0.65
     arrived
    0.63
     approached
    0.61
     exh
    0.61
     confronted
    0.61
    pha
    0.60
     encountered
    0.60
    Activations Density 0.097%

    No Known Activations