© 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. 3-RES-JB
    11. 22722
    Prev
    Next
    INDEX
    Explanations

    mentions of technology-related terms or concepts

    oai_token-act-pair · gpt-3.5-turbo

    terms related to technology and ideological discussions surrounding it

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    CV
    -0.73
     Royale
    -0.72
     Acid
    -0.71
     Sacrament
    -0.70
     Edison
    -0.67
     Grind
    -0.66
     Rite
    -0.66
     Proposition
    -0.64
     Temple
    -0.64
     Stain
    -0.64
    POSITIVE LOGITS
    ths
    0.90
    icip
    0.88
    unci
    0.87
    stration
    0.86
    unciation
    0.85
    etime
    0.85
    cr
    0.83
     techn
    0.80
    iat
    0.80
    oling
    0.79
    Activations Density 0.021%

    No Known Activations