© 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. 1-RES-JB
    11. 20709
    Prev
    Next
    INDEX
    Explanations

    the word 'idd'

    oai_token-act-pair · gpt-3.5-turbo

    variations of the word "id."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     antidepressants
    -0.68
     thin
    -0.67
     tranqu
    -0.67
     metast
    -0.65
     fractures
    -0.62
     Bast
    -0.62
     isolate
    -0.62
    DEM
    -0.61
     Virgin
    -0.57
    MENT
    -0.57
    POSITIVE LOGITS
    idd
    1.40
    olph
    0.96
    ick
    0.94
    itional
    0.94
    lesh
    0.93
    eenth
    0.93
    halla
    0.92
    ety
    0.88
    ilton
    0.86
    augh
    0.86
    Activations Density 0.010%

    No Known Activations