© 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. 9418
    Prev
    Next
    INDEX
    Explanations

    mentions of the concept of "hood", which appears to be related to locations, states, or states of being

    oai_token-act-pair · gpt-3.5-turbo

    references to various forms of "hood," indicating states or conditions related to identity and social roles

    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
    ERC
    -0.81
     FDA
    -0.66
     NIH
    -0.65
     DEF
    -0.64
     Instr
    -0.63
    âķIJâķIJ
    -0.61
    âĸ¬âĸ¬
    -0.60
    çĶ°
    -0.60
     Recommended
    -0.60
     Burk
    -0.59
    POSITIVE LOGITS
    hood
    1.47
    stones
    1.03
    mares
    0.85
    cycles
    0.82
    lord
    0.82
    oom
    0.81
    velt
    0.81
    rants
    0.81
    thood
    0.80
    quartered
    0.79
    Activations Density 0.007%

    No Known Activations