© 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. 11-RES-JB
    11. 9475
    Prev
    Next
    INDEX
    Explanations

    special characters or symbols, particularly slashes followed by numbers

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of slashes

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     square
    -0.84
     breed
    -0.83
     rebuilt
    -0.79
     swell
    -0.78
     densely
    -0.77
     lifetime
    -0.76
     geared
    -0.75
     enclosed
    -0.75
     scratch
    -0.75
     squid
    -0.74
    POSITIVE LOGITS
    whatever
    1.72
    etc
    1.69
    trans
    1.43
    deb
    1.40
    anti
    1.39
    coll
    1.38
    acqu
    1.38
    non
    1.38
    dist
    1.38
    super
    1.37
    Activations Density 0.059%

    No Known Activations