© 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. 7-RES-JB
    11. 13520
    Prev
    Next
    INDEX
    Explanations

    words related to items being untied or unraveled

    oai_token-act-pair · gpt-3.5-turbo

    references to the concept of "looseness" or "loose ends."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    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
    issance
    -0.85
    ãĤ´ãĥ³
    -0.77
    aurus
    -0.77
    illian
    -0.74
    oleon
    -0.72
    igraph
    -0.72
    IENCE
    -0.72
    amus
    -0.71
    ococ
    -0.70
    aur
    -0.70
    POSITIVE LOGITS
    leaf
    1.13
    weed
    0.89
    nesses
    0.86
    ness
    0.82
    cloth
    0.80
     cannon
    0.79
    fitting
    0.79
     leaf
    0.78
    tail
    0.77
    tails
    0.77
    Activations Density 0.012%

    No Known Activations