Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 24576
    5. 8-RES_FS24576-JB
    6. 21798
    Prev
    Next
    INDEX
    Explanations

    numbers and symbols used in code

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs24576-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576
    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

    Neuron Alignment
    Index
    Value
    % of L₁
    364
    +0.10
    0.5%
    405
    +0.10
    0.5%
    433
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    433
    +0.10
    0.03
    649
    +0.10
    0.02
    159
    +0.10
    0.02
    Negative Logits
    wcs
    -0.69
     compr
    -0.65
    milo
    -0.64
    fman
    -0.62
    ometimes
    -0.62
    alties
    -0.62
     ,"
    -0.61
     poppy
    -0.60
    utical
    -0.60
    lees
    -0.59
    POSITIVE LOGITS
    ]
    1.91
    ]"
    1.75
    ?]
    1.64
    ]:
    1.61
    ],
    1.61
    ][
    1.59
    ])
    1.58
    ].
    1.56
    ']
    1.52
    ]).
    1.52
    Activations Density 0.066%

    No Known Activations