INDEX

Explanations

words related to military and government activities

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_fs1536-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

1,536

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

 derby

-0.74

 concentration

-0.74

 steroids

-0.70

orate

-0.70

 reserves

-0.68

 emancipation

-0.67

 numbering

-0.66

 admin

-0.66

 furn

-0.66

 administr

-0.65

POSITIVE LOGITS

Quote

1.24

"...

1.19

"âĢ¦

1.15

http

1.13

https

1.09

Dear

1.08

Feature

1.04

<?

0.99

Looks

0.97

Congratulations

0.96

Activations Density 1.324%

words related to military and government activities

No Comments

No Known Activations