INDEX

Explanations

words and phrases related to discrimination and civil rights issues

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

Adds

-0.71

notes

-0.69

TOR

-0.69

VIDEO

-0.67

ures

-0.67

 Bild

-0.65

 Thumbnails

-0.64

zyme

-0.64

hran

-0.63

links

-0.63

POSITIVE LOGITS

 against

0.96

rimination

0.95

 discrimination

0.92

 discriminating

0.86

 retaliation

0.86

 discriminate

0.85

 discriminated

0.82

 prejudice

0.81

 retali

0.79

 favoring

0.79

Activations Density 0.036%

words and phrases related to discrimination and civil rights issues

No Comments

No Known Activations