INDEX

Explanations

phrases related to political discussions and events

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Canaver

-0.71

ocamp

-0.65

 Mell

-0.64

ATIONAL

-0.64

©¶æ

-0.62

 Proposition

-0.62

 Clancy

-0.61

ģĸ

-0.61

umbn

-0.59

 Lauder

-0.59

POSITIVE LOGITS

 doesnt

1.13

 becomes

1.06

 notwithstanding

1.06

ain

1.01

 wouldn

1.00

 will

0.99

 would

0.99

?).

0.98

 tends

0.98

 alike

0.98

Activations Density 2.290%

phrases related to political discussions and events

No Comments

No Known Activations