INDEX

Explanations

expressions of gratitude towards others

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_fs24576-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

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

 dominates

-0.70

ths

-0.70

Course

-0.68

thing

-0.63

 Worse

-0.63

Temperature

-0.60

ighed

-0.59

hibition

-0.59

POL

-0.59

isation

-0.58

POSITIVE LOGITS

omever

0.86

RIP

0.81

 contributors

0.74

 involved

0.73

Ü

0.73

rats

0.73

involved

0.71

 congr

0.70

 listeners

0.70

 readers

0.69

Activations Density 0.163%

expressions of gratitude towards others

No Comments

No Known Activations