INDEX

Explanations

violence-related vocabulary, specifically related to physical altercations and conflicts

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

@@

-0.60

 Albania

-0.60

fax

-0.59

iac

-0.58

zed

-0.56

 faithfully

-0.56

 Laksh

-0.56

soDeliveryDate

-0.55

phony

-0.54

ItemImage

-0.53

POSITIVE LOGITS

 forefront

1.14

 extent

1.13

 brink

1.11

 fullest

1.04

 detriment

1.02

 nearest

1.00

 lowest

0.96

 verge

0.93

 depths

0.91

 tune

0.91

Activations Density 0.129%

violence-related vocabulary, specifically related to physical altercations and conflicts

No Comments

No Known Activations