Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 11-RES-JB
    6. 19192
    Prev
    Next
    INDEX
    Explanations

    temporal references, such as specific years or time periods

    oai_token-act-pair · gpt-3.5-turbo

    references to temporal concepts and personal experiences

    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
    issors
    -0.79
    =-=-=-=-=-=-=-=-
    -0.74
    nce
    -0.73
    omy
    -0.71
    LOD
    -0.71
    regate
    -0.69
     Volcano
    -0.69
    imation
    -0.69
    acly
    -0.69
    ancial
    -0.68
    POSITIVE LOGITS
     unwilling
    1.65
     unaware
    1.58
     addicted
    1.58
     accustomed
    1.58
     afraid
    1.56
     dissatisfied
    1.55
     aware
    1.55
     unable
    1.54
     ashamed
    1.54
     fearful
    1.52
    Activations Density 0.565%

    No Known Activations