© 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. 8-RES-JB
    11. 7651
    Prev
    Next
    INDEX
    Explanations

    numeric values and units of measurement

    oai_token-act-pair · gpt-3.5-turbo

    instances of examples or case studies

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    acs
    -0.66
    Ĥª
    -0.64
    Assembly
    -0.63
     Grayson
    -0.61
    medium
    -0.61
     Assignment
    -0.60
     PRODUCT
    -0.59
    rus
    -0.59
     Investigator
    -0.59
    BAT
    -0.59
    POSITIVE LOGITS
     famously
    0.89
     likewise
    0.75
     similarly
    0.71
    ß
    0.66
     attest
    0.66
     langu
    0.65
    Þ
    0.65
    vertisement
    0.64
     exting
    0.63
    ñ
    0.61
    Activations Density 1.182%

    No Known Activations