INDEX

Explanations

terms related to sexual abuse, allegations, and assault

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_fs3072-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

3,072

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

travel

-0.87

peror

-0.80

izen

-0.73

views

-0.67

ocular

-0.67

iasm

-0.66

rian

-0.65

AY

-0.65

uber

-0.65

vision

-0.65

POSITIVE LOGITS

 perpetrated

1.15

 victims

1.10

 Victim

0.98

 victim

0.95

 Victims

0.94

 Rape

0.92

 survivors

0.90

 raped

0.90

 inflicted

0.90

 rape

0.87

Activations Density 0.971%

terms related to sexual abuse, allegations, and assault

No Comments

No Known Activations