INDEX

Explanations

mentions of discussions, advancements, science, controversy, and emotional experiences in various contexts

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

 reper

-0.67

 Nare

-0.62

adv

-0.61

 brim

-0.61

 crawl

-0.61

uana

-0.58

 traged

-0.52

 ACTION

-0.52

ULTS

-0.52

 territ

-0.52

POSITIVE LOGITS

ing

2.70

ed

2.15

edIn

1.59

edly

1.56

ment

1.52

ership

1.52

ING

1.42

ation

1.39

ions

1.39

ments

1.38

Activations Density 2.367%

mentions of discussions, advancements, science, controversy, and emotional experiences in various contexts

No Comments

No Known Activations