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. 16159
    Prev
    Next
    INDEX
    Explanations

    words related to strong or forceful actions, possibly violence or impact

    oai_token-act-pair · gpt-3.5-turbo

    references to significant achievements or performances in sports

    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
     conve
    -0.72
     privacy
    -0.69
     mirrors
    -0.65
     cond
    -0.65
     Provision
    -0.64
     permitting
    -0.64
     decor
    -0.63
     Celestial
    -0.63
     delegates
    -0.62
     detailing
    -0.61
    POSITIVE LOGITS
    hit
    4.32
    Hit
    1.82
    hitting
    1.77
     hit
    1.66
     Hit
    1.47
     Hits
    1.46
     HIT
    1.36
     hits
    1.32
    shit
    1.23
    hig
    1.19
    Activations Density 0.016%

    No Known Activations