© 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. 10-RES-JB
    11. 24315
    Prev
    Next
    INDEX
    Explanations

    keywords related to official documents, reports, and recommendations

    oai_token-act-pair · gpt-3.5-turbo

    references to formal processes and documentation in a business or legal context

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    lihood
    -0.78
     wheelchair
    -0.71
    ETHOD
    -0.71
    aughs
    -0.66
    lich
    -0.66
    ãĥ¤
    -0.63
    pse
    -0.61
    odor
    -0.60
    ificate
    -0.59
    Õ
    -0.59
    POSITIVE LOGITS
     ranging
    1.04
     emanating
    1.01
    cape
    0.94
     pertaining
    0.94
     belonging
    0.93
     undertaken
    0.90
     surrounding
    0.89
    chool
    0.89
     collected
    0.89
     relating
    0.89
    Activations Density 0.451%

    No Known Activations