© 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. 17972
    Prev
    Next
    INDEX
    Explanations

    wishes stated in various contexts

    oai_token-act-pair · gpt-3.5-turbo

    the word "wish" and its variations, indicating desires or hopes

    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
    due
    -0.71
    advertisement
    -0.64
    ulative
    -0.63
    helle
    -0.63
    iph
    -0.61
    ulp
    -0.61
    issues
    -0.60
    entin
    -0.59
    ulate
    -0.58
    advant
    -0.58
    POSITIVE LOGITS
     wish
    1.05
     wishes
    1.04
     wished
    0.96
    mares
    0.96
     Lumpur
    0.86
     Pwr
    0.85
    bringer
    0.84
     wishing
    0.83
    igue
    0.79
    ful
    0.76
    Activations Density 0.008%

    No Known Activations