© 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. 4-RES-JB
    11. 6163
    Prev
    Next
    INDEX
    Explanations

    upcoming events or appointments

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "due" indicating scheduled events or obligations

    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
    anon
    -0.87
     Column
    -0.74
    =-=-=-=-=-=-=-=-
    -0.70
     Tornado
    -0.67
    Meg
    -0.64
     sheep
    -0.64
    Co
    -0.64
    agate
    -0.63
    whatever
    -0.63
     Sheep
    -0.62
    POSITIVE LOGITS
     diligence
    0.96
    lling
    0.85
     cancell
    0.77
    */(
    0.77
     deadlines
    0.76
    ority
    0.74
     releg
    0.73
    iciary
    0.73
     ration
    0.72
    giving
    0.71
    Activations Density 0.016%

    No Known Activations