INDEX

Explanations

mentions of specific periods of time

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

ETS

-0.80

quote

-0.75

cause

-0.71

channelAvailability

-0.70

interstitial

-0.66

Area

-0.66

ãĥ¯

-0.64

JV

-0.64

soDeliveryDate

-0.64

FOR

-0.63

POSITIVE LOGITS

 unsuccessful

1.03

 hiatus

1.00

 failed

0.96

 setbacks

0.93

 breakup

0.93

 stint

0.90

 successful

0.88

 debacle

0.88

 initial

0.88

 disastrous

0.88

Activations Density 1.530%

mentions of specific periods of time

No Comments

No Known Activations