© 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. 0-RES-JB
    11. 16395
    Prev
    Next
    INDEX
    Explanations

    phrases related to collaborative efforts or agreements

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    cens
    -0.79
     Wilde
    -0.77
    hin
    -0.77
     Summers
    -0.73
    Settings
    -0.71
    landers
    -0.70
     atheism
    -0.68
     Archangel
    -0.67
    hunt
    -0.66
    Unknown
    -0.64
    POSITIVE LOGITS
     joint
    4.06
     Joint
    2.49
     joints
    2.40
     jointly
    1.85
     partnership
    1.20
     bilateral
    1.16
     junction
    1.13
     coordination
    1.09
     collaboration
    1.00
     sole
    1.00
    Activations Density 0.003%

    No Known Activations