Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsBlog/PodcastSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlog/PodcastGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 5-RES-JB
    6. 6329
    Prev
    Next
    INDEX
    Explanations

    mentions of creative work and related content such as storylines, artwork, and webpages

    oai_token-act-pair · gpt-3.5-turbo

    terms related to narrative elements in visual media, particularly artwork and storylines

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 5-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    5
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    idan
    -0.92
    por
    -0.78
    vik
    -0.76
    val
    -0.75
    ney
    -0.73
    nered
    -0.72
    gas
    -0.71
    cho
    -0.69
    rams
    -0.68
    gans
    -0.68
    POSITIVE LOGITS
     Pwr
    0.96
    VERTISEMENT
    0.77
    ï¸ı
    0.72
    SourceFile
    0.70
    CLASSIFIED
    0.68
    GGGGGGGG
    0.68
    INGTON
    0.67
     APR
    0.67
    soever
    0.66
    ILLE
    0.66
    Activations Density 0.106%

    No Known Activations