© 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. 1-RES-JB
    11. 9516
    Prev
    Next
    INDEX
    Explanations

    different layers of objects or concepts

    oai_token-act-pair · gpt-3.5-turbo

    concepts related to "layers."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Predators
    -0.81
    swick
    -0.76
     Spe
    -0.74
    eve
    -0.72
    ãĥīãĥ©
    -0.71
    cgi
    -0.70
     Wand
    -0.69
    sburgh
    -0.69
    date
    -0.68
    shire
    -0.67
    POSITIVE LOGITS
     layers
    1.74
    layer
    1.29
     layer
    1.21
     layered
    1.16
    ayers
    1.07
    avorite
    1.06
     misunder
    0.94
     Layer
    0.93
    Layer
    0.92
     opacity
    0.90
    Activations Density 0.005%

    No Known Activations