© 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. 11-RES-JB
    11. 21509
    Prev
    Next
    INDEX
    Explanations

    phrases indicating preferences or choices

    oai_token-act-pair · gpt-3.5-turbo

    phrases that express preferences or choices

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    again
    -0.67
    ACP
    -0.67
    xy
    -0.66
    oward
    -0.65
    isal
    -0.65
    brance
    -0.64
    ãħĭ
    -0.63
    RANT
    -0.63
    isner
    -0.63
    Synopsis
    -0.63
    POSITIVE LOGITS
     than
    1.02
    than
    0.82
     Than
    0.80
     avoided
    0.79
     avoid
    0.73
     starve
    0.72
     waited
    0.72
    leave
    0.69
     avoiding
    0.68
     not
    0.64
    Activations Density 0.122%

    No Known Activations