INDEX

Explanations

terms related to political control, governance structures, and legislative issues

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs768-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_768

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

768

Data Type

torch.float32

Hook Point

blocks.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

RAY

-0.73

ennett

-0.72

.;

-0.69

isks

-0.66

onew

-0.65

rison

-0.65

ainers

-0.65

ortun

-0.63

 Solution

-0.62

 Saying

-0.61

POSITIVE LOGITS

 otherwise

0.84

 incidentally

0.82

 antit

0.81

 previously

0.79

 hallmark

0.76

 commonly

0.76

 ordinarily

0.76

 hitherto

0.75

 traditionally

0.75

 anat

0.74

Activations Density 6.681%

terms related to political control, governance structures, and legislative issues

No Comments

No Known Activations