INDEX

Explanations

instances of negation

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ORGE

-0.86

ãĥĵ

-0.81

ãĥ³ãĤ¸

-0.79

ãĤ¤ãĥĪ

-0.78

ãĤ¶

-0.77

ãĥª

-0.71

ãĥ¼ãĥĨãĤ£

-0.71

NetMessage

-0.70

ãĥĥãĥī

-0.70

emonium

-0.69

POSITIVE LOGITS

 already

1.02

 comply

1.00

 succeed

0.96

hin

0.94

 cooperate

0.88

 agree

0.85

 resolve

0.85

 suffice

0.78

 conform

0.77

 complying

0.76

Activations Density 0.099%

instances of negation

No Comments

No Known Activations