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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsBlogSlackPrivacy & 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. 11-RES-JB
    6. 3196
    Prev
    Next
    INDEX
    Explanations

    mentions of law enforcement, crime, and legal action

    oai_token-act-pair · gpt-3.5-turbo

    instances of capital letters at the beginning of sentences or proper nouns

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     pony
    -0.90
    âĢķ
    -0.88
     ðŁĻĤ
    -0.81
     âĢķ
    -0.79
     ______
    -0.77
     blah
    -0.77
     Pony
    -0.77
     Vault
    -0.75
     Sponge
    -0.73
     anyways
    -0.73
    POSITIVE LOGITS
    ensions
    1.41
    earing
    1.32
    ailing
    1.25
    umerous
    1.24
    itting
    1.23
    earchers
    1.22
    iting
    1.22
    ielding
    1.21
    itled
    1.20
    ressed
    1.19
    Activations Density 0.151%

    No Known Activations