© 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. 2679
    Prev
    Next
    INDEX
    Explanations

    phrases indicating variability or dependency on certain conditions

    oai_token-act-pair · gpt-3.5-turbo

    phrases that indicate conditions or factors that are contingent on other variables

    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
    vision
    -0.72
    IU
    -0.70
    talk
    -0.70
    ãĥĵ
    -0.68
    blue
    -0.66
    code
    -0.66
    conduct
    -0.65
    Ò
    -0.65
    ulton
    -0.65
    tha
    -0.65
    POSITIVE LOGITS
     upon
    0.82
     degrees
    0.79
     amounts
    0.76
     amount
    0.74
    encies
    0.72
     depending
    0.70
     conting
    0.70
     caster
    0.68
     subscrib
    0.68
     disproportion
    0.65
    Activations Density 0.020%

    No Known Activations