© 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. 4-RES-JB
    11. 7029
    Prev
    Next
    INDEX
    Explanations

    the word 'yr' with different activation levels

    oai_token-act-pair · gpt-3.5-turbo

    references to years

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

    No Comments

    Negative Logits
     Ou
    -0.81
     Fiesta
    -0.72
    leeve
    -0.71
     Ic
    -0.70
     Shack
    -0.68
     Papua
    -0.68
     Bloom
    -0.66
    ļé
    -0.66
     Canal
    -0.65
     Villa
    -0.65
    POSITIVE LOGITS
    rha
    1.27
    interstitial
    0.93
    rr
    0.90
    rh
    0.90
    andom
    0.85
    annis
    0.83
    umph
    0.82
    azines
    0.82
    acial
    0.80
    rocal
    0.79
    Activations Density 0.009%

    No Known Activations