© 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. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 0-RES-JB
    11. 15707
    Prev
    Next
    INDEX
    Explanations

    fractions and percentages specified with ordinal numbers

    oai_token-act-pair · gpt-3.5-turbo

    fractions and divisions in numerical context

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     res
    -0.74
     dry
    -0.74
     weather
    -0.71
     ballistic
    -0.69
     Des
    -0.67
     surf
    -0.64
     brutally
    -0.64
     values
    -0.64
     sand
    -0.64
     hilar
    -0.62
    POSITIVE LOGITS
    third
    3.66
    fourth
    2.87
    fifth
    2.82
    Third
    2.02
    second
    1.81
    thirds
    1.79
    Fourth
    1.74
    half
    1.68
    quarter
    1.60
    six
    1.45
    Activations Density 0.014%

    No Known Activations