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. 4-RES-JB
    6. 19253
    Prev
    Next
    INDEX
    Explanations

    words related to the programming language and library "Boost"

    oai_token-act-pair · gpt-3.5-turbo

    repetitive phrases or terms related to "Boost" and "BYU"

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

    No Comments

    Negative Logits
    ktop
    -0.76
    opter
    -0.73
     Ips
    -0.70
    DonaldTrump
    -0.69
    ect
    -0.68
    Magikarp
    -0.67
    idan
    -0.67
    aceae
    -0.66
    mediate
    -0.66
    ocene
    -0.65
    POSITIVE LOGITS
     blah
    1.12
    bies
    0.76
    ���
    0.74
     booster
    0.70
    rous
    0.65
    ��
    0.65
    aneers
    0.64
     boosters
    0.64
    ueless
    0.62
    nder
    0.61
    Activations Density 0.011%

    No Known Activations