INDEX

Explanations

references to historical events or controversies, specifically related to crime or extremism

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

iances

-0.80

wcsstore

-0.77

jun

-0.77

erd

-0.74

 mates

-0.72

gary

-0.71

gap

-0.70

tails

-0.70

weeney

-0.67

body

-0.66

POSITIVE LOGITS

ocaust

1.25

 Holocaust

1.06

 survivors

0.98

 survivor

0.97

 Genocide

0.92

 atrocities

0.91

 victims

0.91

 extermination

0.90

 genocide

0.89

 massacre

0.89

Activations Density 0.068%

references to historical events or controversies, specifically related to crime or extremism

No Comments

No Known Activations