INDEX

Explanations

phrases or words related to variations or differences

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_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

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

DEF

-0.76

ãĤ¸

-0.74

iron

-0.69

NN

-0.67

çīĪ

-0.66

ainment

-0.66

lest

-0.65

Otherwise

-0.64

Li

-0.63

WARN

-0.63

POSITIVE LOGITS

 differing

1.21

 different

1.14

 varying

1.10

 depending

1.09

 severity

1.01

 geographically

0.99

 variations

0.98

different

0.97

 variation

0.95

 geographic

0.93

Activations Density 0.250%

phrases or words related to variations or differences

No Comments

No Known Activations