INDEX

Explanations

phrases related to trips, tours, visits, and delegations

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_fs1536-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

1,536

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

rity

-0.89

corn

-0.78

rate

-0.69

rates

-0.67

reme

-0.67

 procedural

-0.64

rator

-0.64

omething

-0.63

 Proced

-0.62

bra

-0.61

POSITIVE LOGITS

 itiner

1.19

 abroad

1.06

 Tours

1.04

 destinations

0.94

 tours

0.93

 tour

0.92

 touring

0.91

 visiting

0.87

 Disneyland

0.84

 trips

0.82

Activations Density 2.323%

phrases related to trips, tours, visits, and delegations

No Comments

No Known Activations