INDEX

Explanations

phrases related to actions and decisions made forcefully, unconventionally, or critically

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

uncture

-0.70

Card

-0.65

 Returning

-0.64

rice

-0.62

 Pend

-0.61

Benef

-0.61

ilies

-0.59

etta

-0.57

æĿ

-0.57

ection

-0.57

POSITIVE LOGITS

 lengths

0.77

 unnoticed

0.77

 sidx

0.73

 overboard

0.72

 hiatus

0.70

 shopping

0.67

 baths

0.65

Ô

0.65

zag

0.63

inki

0.62

Activations Density 0.836%

phrases related to actions and decisions made forcefully, unconventionally, or critically

No Comments

No Known Activations