© 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. 24111
    Prev
    Next
    INDEX
    Explanations

    phrases related to events that have occurred previously

    oai_token-act-pair · gpt-3.5-turbo

    instances of the phrase "the first time."

    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
    Abs
    -0.67
    Hyper
    -0.64
    avoid
    -0.63
    mem
    -0.62
    ARM
    -0.62
    Exit
    -0.62
    android
    -0.61
    peace
    -0.61
    Maximum
    -0.61
    ————————————————
    -0.60
    POSITIVE LOGITS
     anymore
    0.88
     slightest
    0.83
     nor
    0.78
     baseman
    0.73
     sighting
    0.73
     volley
    0.72
     responders
    0.72
     rode
    0.71
     accus
    0.69
     holiest
    0.69
    Activations Density 0.031%

    No Known Activations