INDEX

Explanations

repeated punctuation or structural markers in text

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs49152-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

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

 sprung

-0.80

-0.73

æ©

-0.68

ãĥ¼ãĥĨ

-0.64

*/(

-0.64

ciation

-0.64

enta

-0.63

 overd

-0.63

 nodd

-0.63

bledon

-0.61

POSITIVE LOGITS

1.27

._

1.03

..

1.02

.)

0.94

*.

0.94

."

0.90

-.

0.86

0.85

(.

0.81

..................

0.80

Activations Density 0.003%

repeated punctuation or structural markers in text

No Comments

No Known Activations