INDEX

Explanations

mentions of reading or books

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_fs24576-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

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

xon

-0.67

Ã±o

-0.64

Enlarge

-0.62

lla

-0.61

ortality

-0.60

ascal

-0.59

ctions

-0.59

cker

-0.59

trl

-0.58

ality

-0.58

POSITIVE LOGITS

 aloud

1.49

just

1.15

 comprehension

1.09

dress

0.98

 excerpts

0.96

 books

0.94

write

0.89

 texts

0.88

mitt

0.88

mill

0.86

Activations Density 0.040%

mentions of reading or books

No Comments

No Known Activations