© 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. 9003
    Prev
    Next
    INDEX
    Explanations

    phrases or sentences that convey a point of comparison or contrast

    oai_token-act-pair · gpt-3.5-turbo

    instances of the term "contrast" and its use in comparing different subjects or situations

    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
    otin
    -0.78
    zyme
    -0.78
    iston
    -0.75
    authorized
    -0.73
    vae
    -0.71
    enium
    -0.70
    atom
    -0.69
    cers
    -0.68
    thing
    -0.66
    aptic
    -0.66
    POSITIVE LOGITS
     contrasts
    0.97
     contrast
    0.88
     juxtap
    0.87
     Contrast
    0.84
    lihood
    0.77
     GOODMAN
    0.76
     contrasted
    0.75
    xual
    0.72
    isons
    0.72
     contrasting
    0.71
    Activations Density 0.005%

    No Known Activations