INDEX

Explanations

technical terms related to programming, computing, and mathematical notations

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

ogi

-0.75

afety

-0.73

ariat

-0.72

jon

-0.72

intent

-0.71

imum

-0.70

mys

-0.70

ables

-0.69

ubi

-0.69

obal

-0.69

POSITIVE LOGITS

 transsexual

0.97

 converter

0.91

 alike

0.89

 comparisons

0.87

 correspondence

0.86

 ratio

0.79

 dich

0.77

 conver

0.77

 bilingual

0.76

 ratios

0.72

Activations Density 1.011%

technical terms related to programming, computing, and mathematical notations

No Comments

No Known Activations