INDEX

Explanations

phrases related to political events and government policies

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

nels

-0.80

ording

-0.78

xual

-0.73

object

-0.72

ĳå£«

-0.70

flies

-0.70

Kind

-0.69

event

-0.69

!/

-0.68

JECT

-0.68

POSITIVE LOGITS

 largest

1.18

 poorest

1.05

 wealthiest

1.03

 finest

1.00

 tallest

1.00

 premier

1.00

 richest

0.98

 longest

0.97

 biggest

0.97

 newest

0.95

Activations Density 0.687%

phrases related to political events and government policies

No Comments

No Known Activations