© 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. 6271
    Prev
    Next
    INDEX
    Explanations

    captions or titles

    oai_token-act-pair · gpt-3.5-turbo

    instances of captions related to images or media content

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    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
    ois
    -0.67
    rush
    -0.65
     troll
    -0.65
    side
    -0.64
     incomp
    -0.62
     flat
    -0.58
     opposed
    -0.58
     deny
    -0.58
     bi
    -0.57
     dig
    -0.57
    POSITIVE LOGITS
     Caption
    5.07
     caption
    1.34
     largeDownload
    1.27
     Thumbnails
    1.19
    thumbnails
    1.07
    Interstitial
    0.99
    OTOS
    0.95
     Photographer
    0.95
     Advertisement
    0.92
    utterstock
    0.91
    Activations Density 0.015%

    No Known Activations