INDEX

Explanations

technical terms related to software tools and development

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

 Sons

-0.72

ŃĶ

-0.71

igi

-0.71

 Alive

-0.63

 Chapters

-0.62

oul

-0.62

enda

-0.61

 Chapter

-0.60

 gloom

-0.59

uras

-0.59

POSITIVE LOGITS

kit

1.31

tips

1.21

sonian

1.09

 tools

1.02

 tool

1.00

assisted

0.96

tip

0.94

ologies

0.92

tools

0.85

tool

0.83

Activations Density 1.054%

technical terms related to software tools and development

No Comments

No Known Activations