© 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. 3-RES-JB
    11. 8394
    Prev
    Next
    INDEX
    Explanations

    words related to sports rotations, including players being in or put into rotation

    oai_token-act-pair · gpt-3.5-turbo

    references to organizational or positional sequences

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    externalActionCode
    -0.79
    erry
    -0.73
     Asians
    -0.70
    Obama
    -0.69
    native
    -0.69
    omething
    -0.68
    CHA
    -0.68
    umble
    -0.68
    asper
    -0.67
    eteria
    -0.66
    POSITIVE LOGITS
     rotation
    1.13
     rotating
    0.84
     rotated
    0.82
     rotate
    0.78
     destro
    0.77
     mund
    0.75
    ular
    0.74
    otation
    0.74
     slic
    0.73
     revolves
    0.72
    Activations Density 0.006%

    No Known Activations