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. 8775
    Prev
    Next
    INDEX
    Explanations

    words and phrases related to rapid changes or strong increases in a negative context

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "sharply" and its usage in various contexts

    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
    rity
    -0.80
    busters
    -0.76
    orno
    -0.75
    gres
    -0.71
    itu
    -0.70
    acity
    -0.70
    ador
    -0.69
    ourke
    -0.68
    ocene
    -0.68
    allas
    -0.66
    POSITIVE LOGITS
     diver
    0.84
    ãĥ£
    0.82
     tuned
    0.80
     downwards
    0.77
     downward
    0.73
     contrad
    0.72
     upward
    0.71
     differentiated
    0.71
     deline
    0.71
     sharply
    0.70
    Activations Density 0.009%

    No Known Activations