INDEX

Explanations

phrases indicating certainty or emphasis

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

ourses

-0.71

 Written

-0.66

aired

-0.66

tumblr

-0.65

Course

-0.64

urated

-0.62

 decom

-0.60

hung

-0.58

und

-0.58

ieg

-0.57

POSITIVE LOGITS

 though

0.79

0.74

:]

0.73

lycer

0.72

âĶĢ

0.66

 emphatically

0.64

!:

0.64

 ABOUT

0.63

Kev

0.63

:,

0.63

Activations Density 1.411%

phrases indicating certainty or emphasis

No Comments

No Known Activations