INDEX

Explanations

words related to positive emotions or expressions, particularly those that evoke joy or happiness

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs49152-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

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

*/(

-0.85

 Nightmares

-0.76

 Hole

-0.66

 cider

-0.65

 Jonah

-0.64

 Forth

-0.63

 Ness

-0.61

 Myster

-0.61

 Skydragon

-0.61

kus

-0.60

POSITIVE LOGITS

«ĺ

1.27

¢

1.15

¬¼

1.11

ĺ

1.05

Ķ

1.02

¦

1.02

Ŀ

0.99

Ĥª

0.98

ľ

0.97

Ļ½

0.97

Activations Density 0.002%

words related to positive emotions or expressions, particularly those that evoke joy or happiness

No Comments

No Known Activations