INDEX

Explanations

references to specific historical events and information

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

 envy

-0.73

 Cosponsors

-0.70

ãĥ³ãĤ¸

-0.67

<?

-0.66

 millenn

-0.63

²¾

-0.62

nic

-0.60

 aliases

-0.58

mong

-0.58

WER

-0.57

POSITIVE LOGITS

 expires

0.98

uberty

0.87

 landfall

0.87

 arrives

0.85

 arrive

0.81

 expire

0.79

pection

0.76

apses

0.74

pleted

0.70

 arrived

0.70

Activations Density 2.364%

references to specific historical events and information

No Comments

No Known Activations