© 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
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 4-RES-JB
    6. 15303
    Prev
    Next
    INDEX
    Explanations

    text related to captions or descriptions of photographs

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "Photo" followed by numeric values, indicating the presence of images or pictures

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

    No Comments

    Negative Logits
    uckle
    -0.83
    ensable
    -0.79
    astically
    -0.76
    anche
    -0.74
    Ͻ
    -0.73
    resses
    -0.72
    ims
    -0.72
    unic
    -0.72
    powers
    -0.72
    etimes
    -0.70
    POSITIVE LOGITS
     Photo
    0.96
     illustration
    0.89
     Album
    0.89
     Courtesy
    0.89
    journal
    0.88
    Photo
    0.87
     Photograph
    0.86
     Gallery
    0.84
     Editing
    0.83
     Photographer
    0.81
    Activations Density 0.020%

    No Known Activations