© 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. 23469
    Prev
    Next
    INDEX
    Explanations

    phrases related to internal mechanisms or operations

    oai_token-act-pair · gpt-3.5-turbo

    references to inner structures or systems, particularly in a societal context

    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
    atoes
    -0.84
    orthy
    -0.83
    eday
    -0.83
    essors
    -0.78
    dayName
    -0.75
    ORK
    -0.71
    enegger
    -0.71
    icable
    -0.70
    iculty
    -0.69
    ILLE
    -0.68
    POSITIVE LOGITS
     workings
    1.34
    most
    1.30
     combustion
    0.89
     sanct
    0.87
     circle
    0.79
     turmoil
    0.76
    wear
    0.76
    ranean
    0.76
    HTML
    0.74
     combust
    0.73
    Activations Density 0.022%

    No Known Activations