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. 0-RES-JB
    6. 3539
    Prev
    Next
    INDEX
    Explanations

    phrases related to catastrophic events or tragedies

    oai_token-act-pair · gpt-3.5-turbo

    the presence of the term "Se" and its various repetitions, indicating a strong focus on specific data or metrics

    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
     backwards
    -0.70
     bonding
    -0.70
     oxy
    -0.68
    affer
    -0.68
    regor
    -0.65
     goof
    -0.65
     backward
    -0.64
     trust
    -0.63
     penalty
    -0.62
     disrespect
    -0.61
    POSITIVE LOGITS
    Se
    3.64
     Se
    1.92
    se
    1.67
    SE
    1.50
    Seg
    1.48
     Seek
    1.25
    Cho
    1.25
    Sequ
    1.21
    Sel
    1.19
    Sc
    1.18
    Activations Density 0.014%

    No Known Activations