INDEX

Explanations

references to food items and related topics

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

 Caldwell

-0.74

âĢ¢âĢ¢âĢ¢âĢ¢

-0.72

Jed

-0.71

worthiness

-0.69

 Armory

-0.68

PV

-0.68

 Jenner

-0.68

 Moff

-0.67

 Wink

-0.66

 Trooper

-0.66

POSITIVE LOGITS

 eaten

1.10

 meals

1.06

mone

1.06

eteria

1.05

 foods

0.98

rition

0.97

 nutrit

0.96

meat

0.95

Ingredients

0.95

cake

0.94

Activations Density 0.985%

references to food items and related topics

No Comments

No Known Activations