© 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. 7642
    Prev
    Next
    INDEX
    Explanations

    numeric estimates or statistical values

    oai_token-act-pair · gpt-3.5-turbo

    quantitative estimates and assessments related to time and financial data

    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
    arak
    -0.78
    emi
    -0.78
    Synopsis
    -0.76
    avage
    -0.73
    ère
    -0.73
    photos
    -0.73
    mares
    -0.72
    rim
    -0.72
     Carbuncle
    -0.71
    irez
    -0.71
    POSITIVE LOGITS
     effectiveness
    1.30
     probability
    1.24
     probabilities
    1.22
     likelihood
    1.21
     prevalence
    1.11
     popularity
    1.07
     percentage
    1.06
     accuracy
    1.05
     usefulness
    1.05
     extent
    1.04
    Activations Density 0.316%

    No Known Activations