© 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. 4-RES-JB
    11. 15293
    Prev
    Next
    INDEX
    Explanations

    phrases indicating the speaker's subjective opinion or judgment

    oai_token-act-pair · gpt-3.5-turbo

    expressions of personal opinion or assertion

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

    No Comments

    Negative Logits
    ¥ŀ
    -0.82
    tele
    -0.71
     conflic
    -0.70
    ĸļ士
    -0.70
    phrine
    -0.68
    ritional
    -0.68
     cumbers
    -0.68
    itially
    -0.66
    mental
    -0.65
    atched
    -0.64
    POSITIVE LOGITS
     goodbye
    1.33
     hello
    1.03
     Goodbye
    0.89
    ings
    0.81
     aloud
    0.76
     hi
    0.74
    INGS
    0.71
     farewell
    0.71
    ieu
    0.70
     Merry
    0.69
    Activations Density 0.063%

    No Known Activations