© 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. 0-RES-JB
    11. 16453
    Prev
    Next
    INDEX
    Explanations

    words related to different departments

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    gpu
    -0.74
    ise
    -0.72
    flame
    -0.70
    cest
    -0.67
    ced
    -0.65
    everything
    -0.64
     offline
    -0.62
    nob
    -0.61
    raise
    -0.61
    iser
    -0.61
    POSITIVE LOGITS
     Department
    3.53
    Department
    2.90
     Dept
    2.83
     department
    2.44
     departments
    1.98
     Ministry
    1.90
     Bureau
    1.90
     Office
    1.64
     Secretary
    1.61
     Directorate
    1.49
    Activations Density 0.019%

    No Known Activations