© 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. 0-RES-JB
    11. 1319
    Prev
    Next
    INDEX
    Explanations

    terms related to technical specifications or details

    oai_token-act-pair · gpt-3.5-turbo

    the word "tes" and its variations, indicating a focus on some form of testing or examinations

    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
     Barry
    -0.67
     Portugal
    -0.65
     extradition
    -0.61
     boom
    -0.59
     prevention
    -0.59
     Lisbon
    -0.58
     famine
    -0.58
     Irwin
    -0.58
     judgment
    -0.57
     Fitzgerald
    -0.57
    POSITIVE LOGITS
    tes
    4.98
    te
    2.17
    testers
    1.69
    ta
    1.32
    ts
    1.28
    ttes
    1.27
    test
    1.26
    tests
    1.21
    tis
    1.19
    testing
    1.18
    Activations Density 0.006%

    No Known Activations