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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & 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. 8-RES-JB
    6. 22025
    Prev
    Next
    INDEX
    Explanations

    sentences presenting strong opinions or emphasizing a point

    oai_token-act-pair · gpt-3.5-turbo

    expressions of opinion or assertion

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    nton
    -0.59
    obbies
    -0.59
    culosis
    -0.57
    elight
    -0.56
    oqu
    -0.56
    redited
    -0.56
     Pict
    -0.55
    sequently
    -0.55
    egu
    -0.55
    ospital
    -0.54
    POSITIVE LOGITS
     fucking
    0.71
     goddamn
    0.63
     bullshit
    0.62
    fuck
    0.61
     fucked
    0.60
     Canaver
    0.60
     Voldemort
    0.60
    _>
    0.60
     fuck
    0.59
     physicists
    0.59
    Activations Density 1.721%

    No Known Activations