© 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. 3213
    Prev
    Next
    INDEX
    Explanations

    phrases related to objectivity or the lack thereof

    oai_token-act-pair · gpt-3.5-turbo

    references to objectives or goals

    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
    oles
    -0.82
    sing
    -0.75
    letters
    -0.71
    cling
    -0.69
    paces
    -0.67
    ingers
    -0.67
     Printed
    -0.66
    Kids
    -0.66
     Tycoon
    -0.65
    Fuel
    -0.65
    POSITIVE LOGITS
     objective
    1.23
     observer
    0.99
     Objective
    0.95
    ignty
    0.87
     objectives
    0.87
     goal
    0.82
     guiActiveUn
    0.81
    lehem
    0.80
    ives
    0.79
    ablishment
    0.77
    Activations Density 0.007%

    No Known Activations