© 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. Apollo Research · Taylor · Identifying Functionally Important Features with End-to-End Sparse Dictionary Learning
    5. GPT2-Small
    7. Residual Similar CE Local
    9. 6-RES_SCL-AJT
    11. 40477
    Prev
    Next
    INDEX
    Explanations

    phrases related to informational details and factual statements

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    neuronpedia/gpt2-small__res_scl-ajt/6-res_scl-ajt
    Prompts (Dashboard)
    12,288 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    46,080
    Data Type
    torch.float32
    Hook Point
    blocks.6.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    apollo-research/Skylion007-openwebtext-tokenizer-gpt2
    Hook Point Layer
    6
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    447
    +0.08
    0.7%
    287
    +0.07
    0.6%
    680
    +0.06
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    287
    +0.08
    0.09
    434
    +0.07
    0.09
    696
    +0.06
    0.13
    Negative Logits
     CJ
    -0.34
     ed
    -0.31
     nib
    -0.31
     actionGroup
    -0.31
    oint
    -0.30
     consequ
    -0.30
     PG
    -0.30
     stricken
    -0.30
     cogn
    -0.30
    ointed
    -0.29
    POSITIVE LOGITS
    abouts
    0.50
    ngth
    0.39
    geist
    0.39
    ramid
    0.39
    ket
    0.38
    olver
    0.36
    ldon
    0.36
    Else
    0.36
    abases
    0.35
    velt
    0.35
    Activations Density 9.046%

    No Known Activations