INDEX

Explanations

sporting event-related terms, particularly outcomes like wins and losses

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_fs6144-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_6144

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

6,144

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

pmwiki

-0.99

iscal

-0.83

 Journals

-0.80

erred

-0.78

conserv

-0.76

vec

-0.73

attribute

-0.72

erity

-0.72

rared

-0.71

 curric

-0.71

POSITIVE LOGITS

 lowly

1.01

 Kansas

0.99

 Detroit

0.95

 Tampa

0.94

 Jacksonville

0.94

 Carolina

0.93

 Minnesota

0.91

 Toledo

0.90

 Pittsburgh

0.90

 opponent

0.90

Activations Density 0.507%

sporting event-related terms, particularly outcomes like wins and losses

No Comments

No Known Activations