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. 8761
    Prev
    Next
    INDEX
    Explanations

    descriptions related to a specific style, such as "Soviet-style" or "French-style", in various contexts

    oai_token-act-pair · gpt-3.5-turbo

    references to distinct styles associated with cultural or political 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
    ora
    -0.88
    umi
    -0.83
    teenth
    -0.79
     McKay
    -0.73
    unes
    -0.72
    oras
    -0.71
    orb
    -0.68
    aman
    -0.68
    ÃŃs
    -0.68
    duct
    -0.67
    POSITIVE LOGITS
    lihood
    0.97
    heet
    0.79
    style
    0.71
     GP
    0.68
     Dragonbound
    0.67
     prevail
    0.65
     style
    0.64
     executions
    0.63
     sheets
    0.62
     consistency
    0.62
    Activations Density 0.014%

    No Known Activations