© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 4-RES-JB
    11. 19417
    Prev
    Next
    INDEX
    Explanations

    adjectives describing unique or characteristic features

    oai_token-act-pair · gpt-3.5-turbo

    terms that describe unique attributes or qualities

    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
    ©¶æ
    -0.79
    wives
    -0.77
    Ö¼
    -0.76
    heed
    -0.75
    ergy
    -0.75
    ationally
    -0.72
    hens
    -0.72
    à¥
    -0.68
    den
    -0.67
    ser
    -0.67
    POSITIVE LOGITS
    eatures
    0.99
     distinctive
    0.88
     traits
    0.87
     characteristics
    0.82
     qualities
    0.80
     deviations
    0.79
     remem
    0.77
     characteristic
    0.75
     charact
    0.73
     Features
    0.73
    Activations Density 0.019%

    No Known Activations