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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 3453
    Prev
    Next
    INDEX
    Explanations

    instances where possessive pronouns (e.g., his, her) are used

    oai_token-act-pair · gpt-3.5-turbo

    frequent use of the word "the."

    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
    hots
    -0.91
    ibaba
    -0.89
    Europe
    -0.83
    peak
    -0.77
    Indust
    -0.77
    english
    -0.75
    iversity
    -0.72
    udos
    -0.72
    osphere
    -0.71
    ollo
    -0.71
    POSITIVE LOGITS
     couple
    1.30
     youngster
    1.28
     toddler
    1.23
     boy
    1.23
     eldest
    1.23
     girl
    1.21
     teenager
    1.21
     family
    1.20
     twins
    1.12
     youngest
    1.11
    Activations Density 0.311%

    No Known Activations