© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 7-RES-JB
    11. 3136
    Prev
    Next
    INDEX
    Explanations

    instances where a specific word "which" is used followed by the number 9 or 10

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 7-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    7
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    politics
    -0.78
    let
    -0.75
    LET
    -0.68
    Calling
    -0.62
    CLOSE
    -0.62
     Calling
    -0.61
    UGH
    -0.60
     dividing
    -0.60
     RTX
    -0.59
    UG
    -0.59
    POSITIVE LOGITS
     originated
    0.85
     resulted
    0.85
     contributed
    0.82
     involve
    0.80
     lasted
    0.79
     derive
    0.78
     consisted
    0.78
     exceeded
    0.77
     specialize
    0.77
     yielded
    0.77
    Activations Density 0.022%

    No Known Activations