© 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. 20173
    Prev
    Next
    INDEX
    Explanations

    Japanese names, primarily with "iko" at the end

    oai_token-act-pair · gpt-3.5-turbo

    mentions of individuals, particularly in a Japanese context

    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
     mistaken
    -0.73
     partnerships
    -0.72
    roads
    -0.68
    drawn
    -0.65
     misunderstanding
    -0.64
    Correct
    -0.63
     marrow
    -0.62
     miscar
    -0.62
     campuses
    -0.62
     reconciliation
    -0.61
    POSITIVE LOGITS
    iko
    1.21
    ichi
    0.94
    omi
    0.93
    imura
    0.86
    itsu
    0.85
    oko
    0.84
    obi
    0.83
    oto
    0.83
    ira
    0.82
    iro
    0.81
    Activations Density 0.007%

    No Known Activations