INDEX

Explanations

words related to emotions, specifically focusing on the feeling of anger

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

glas

-0.89

sites

-0.88

nown

-0.87

 livest

-0.85

akeru

-0.82

aer

-0.78

icrobial

-0.78

ãĥ¯

-0.77

aimon

-0.74

uchin

-0.72

POSITIVE LOGITS

 outburst

1.03

 mobs

0.97

 angry

0.93

 anger

0.85

 indignation

0.84

 angrily

0.81

 frustration

0.77

 Birds

0.76

storms

0.76

 resentment

0.74

Activations Density 0.045%

words related to emotions, specifically focusing on the feeling of anger

No Comments

No Known Activations