© 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. 1274
    Prev
    Next
    INDEX
    Explanations

    the letter "u" when it is repeated

    oai_token-act-pair · gpt-3.5-turbo

    references to the letter 'u'

    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
    arious
    -0.84
    quin
    -0.71
    isSpecialOrderable
    -0.65
     Revel
    -0.64
    SPONSORED
    -0.63
     Citation
    -0.61
    pour
    -0.61
     Hunger
    -0.61
    holm
    -0.60
    hart
    -0.59
    POSITIVE LOGITS
     u
    3.53
     ur
    1.68
     um
    1.22
     U
    1.21
     ul
    1.20
    U
    1.19
     ub
    1.19
     o
    1.19
    u
    1.11
     y
    1.11
    Activations Density 0.011%

    No Known Activations