© 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. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 8-RES-JB
    11. 23613
    Prev
    Next
    INDEX
    Explanations

    numerical data points with a focus on percentages, quantities, and comparisons

    oai_token-act-pair · gpt-3.5-turbo

    numerical data related to points, percentages, and other statistical values

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.hook_resid_pre
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ibaba
    -0.71
    oided
    -0.60
     Pengu
    -0.59
    vana
    -0.59
    rology
    -0.58
    deen
    -0.58
     ling
    -0.56
    iquid
    -0.54
    ovsky
    -0.54
    rera
    -0.53
    POSITIVE LOGITS
    %.
    1.01
     margin
    0.87
     leaps
    0.85
    dB
    0.83
    %,
    0.81
     margins
    0.76
    %"
    0.73
    %;
    0.73
     factor
    0.67
     thirds
    0.67
    Activations Density 0.115%

    No Known Activations