© 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. 4556
    Prev
    Next
    INDEX
    Explanations

    terms or phrases related to the end or conclusion

    oai_token-act-pair · gpt-3.5-turbo

    the end of sentences or paragraphs

    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
    ppo
    -0.80
    IGHTS
    -0.70
    ategory
    -0.68
    ichick
    -0.66
    kaya
    -0.65
     Aval
    -0.62
    CHA
    -0.60
    PDATE
    -0.60
     Byrd
    -0.60
    elsius
    -0.59
    POSITIVE LOGITS
    angered
    1.02
    angering
    0.97
    lich
    0.95
    urance
    0.93
    erer
    0.89
    orph
    0.87
    ocrin
    0.87
    enson
    0.84
    ela
    0.83
    orse
    0.83
    Activations Density 0.015%

    No Known Activations