© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 12408
    Prev
    Next
    INDEX
    Explanations

    references to the company Verizon in texts

    oai_token-act-pair · gpt-3.5-turbo

    references to the brand Verizon

    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
     sweats
    -0.64
     Percy
    -0.63
     Dangerous
    -0.63
     artifacts
    -0.61
     Mo
    -0.61
     bruising
    -0.60
     gruesome
    -0.59
     cult
    -0.59
     rh
    -0.58
     repetitive
    -0.58
    POSITIVE LOGITS
    izon
    5.15
    izons
    2.25
    izontal
    1.20
    izont
    1.19
    awei
    1.13
    IZ
    1.07
    iz
    1.03
     Horizon
    1.00
    itone
    1.00
    ismo
    0.99
    Activations Density 0.029%

    No Known Activations