INDEX

Explanations

quotations and memories related to personal experiences

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

 Chal

-0.62

 Xuan

-0.60

building

-0.59

 Trap

-0.58

Kag

-0.58

SIG

-0.57

è¦ļéĨĴ

-0.57

 Case

-0.57

Vis

-0.57

{*

-0.55

POSITIVE LOGITS

be

1.22

 suffice

0.94

 gladly

0.94

 introduce

0.93

 prefer

0.92

've

0.92

ĸļ

0.92

 allow

0.91

 doubtless

0.90

 have

0.89

Activations Density 0.470%

quotations and memories related to personal experiences

No Comments

No Known Activations