INDEX

Explanations

news-related entities and terms from Reuters and AFP, as well as related actions and events

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_fs3072-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

3,072

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

hal

-0.71

 sediment

-0.69

gone

-0.69

 reinvest

-0.69

gran

-0.68

sed

-0.65

 tranquil

-0.64

gaard

-0.63

 chilling

-0.63

within

-0.60

POSITIVE LOGITS

 Images

1.22

sburg

1.16

Images

0.94

PLIED

0.91

 IMAGES

0.86

sburgh

0.83

photo

0.77

PLE

0.77

TN

0.75

PLA

0.75

Activations Density 0.434%

news-related entities and terms from Reuters and AFP, as well as related actions and events

No Comments

No Known Activations