INDEX

Explanations

statements describing events or actions happening

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

ãĥĨ

-0.73

QB

-0.69

MY

-0.68

çĶ°

-0.68

ãĥĩ

-0.67

Border

-0.65

Ã±a

-0.65

æ°

-0.64

ãĤ©

-0.63

äºĶ

-0.62

POSITIVE LOGITS

 nonetheless

1.82

 nevertheless

1.81

 suffice

1.14

 anecd

1.08

 luckily

0.94

 certainly

0.90

hey

0.90

 fortunately

0.87

etheless

0.85

 anyway

0.83

Activations Density 1.370%

statements describing events or actions happening

No Comments

No Known Activations