© 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. 8-RES-JB
    11. 17133
    Prev
    Next
    INDEX
    Explanations

    dates or acronyms with specific formatting, such as 'year. Now it's just', 'D. or ' . 'IG

    oai_token-act-pair · gpt-3.5-turbo

    unspecified statistics and references

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     interviewer
    -0.62
     vide
    -0.61
     declass
    -0.60
     steering
    -0.58
     broch
    -0.57
     instructional
    -0.57
     drilled
    -0.57
     ped
    -0.56
    inarily
    -0.56
     bounded
    -0.55
    POSITIVE LOGITS
    com
    0.85
    Va
    0.85
    $.
    0.82
    COM
    0.82
    Org
    0.78
    MX
    0.77
    org
    0.76
    é¾įå
    0.75
    CO
    0.74
    mx
    0.73
    Activations Density 0.132%

    No Known Activations