© 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. 0-RES-JB
    11. 23149
    Prev
    Next
    INDEX
    Explanations

    the name "Haji" appearing in various contexts

    oai_token-act-pair · gpt-3.5-turbo

    proper nouns, particularly names or references to specific individuals and places

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     PRO
    -0.63
     rack
    -0.63
     Fab
    -0.62
     LX
    -0.60
     mantle
    -0.59
     glut
    -0.58
     basket
    -0.57
     SEC
    -0.57
     Freeze
    -0.57
     pro
    -0.56
    POSITIVE LOGITS
    aji
    4.63
    shi
    1.15
    oji
    1.01
    aida
    0.98
    inen
    0.97
    ANI
    0.97
    istani
    0.94
     Gujar
    0.94
    aj
    0.90
    ugu
    0.90
    Activations Density 0.022%

    No Known Activations