© 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. 2-RES-JB
    11. 4521
    Prev
    Next
    INDEX
    Explanations

    references to a specific term or concept related to databases or web development

    oai_token-act-pair · gpt-3.5-turbo

    mentions of specific movies, software, or relational concepts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 2-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    2
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Dhabi
    -0.74
    rave
    -0.69
    rier
    -0.67
    istani
    -0.67
     Cath
    -0.65
    RAY
    -0.65
     Citizens
    -0.64
    brush
    -0.64
    IRT
    -0.64
    irtual
    -0.63
    POSITIVE LOGITS
    mys
    0.94
    pace
    0.94
    erver
    0.92
    paces
    0.87
    achine
    0.82
    fell
    0.81
    ogyn
    0.77
    antis
    0.77
    eri
    0.74
    ql
    0.73
    Activations Density 0.019%

    No Known Activations