© 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. 11-RES-JB
    11. 9957
    Prev
    Next
    INDEX
    Explanations

    numerical measurements, specifically focusing on diameters

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "in" used multiple times within a context

    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
    something
    -0.69
    ãĤ´ãĥ³
    -0.68
    oping
    -0.67
    idable
    -0.63
    Winged
    -0.63
    PASS
    -0.61
    think
    -0.61
    ISION
    -0.61
    WHERE
    -0.61
    alist
    -0.60
    POSITIVE LOGITS
     total
    1.20
     diameter
    1.20
    verts
    1.01
    clusively
    1.00
     length
    0.97
     duration
    0.95
     succession
    0.93
     circumference
    0.91
    versions
    0.90
    clusions
    0.87
    Activations Density 0.102%

    No Known Activations