INDEX

Explanations

names related to legal cases or criminal incidents involving perturbation in a neighborhood community

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

unal

-0.78

ggle

-0.74

undai

-0.72

awaru

-0.67

ntil

-0.66

urities

-0.65

arching

-0.64

sembly

-0.64

ccording

-0.63

ablishment

-0.63

POSITIVE LOGITS

ville

0.99

hurst

0.98

gren

0.94

burg

0.94

ston

0.90

 Farms

0.89

Jr

0.88

sey

0.88

burgh

0.88

III

0.86

Activations Density 2.149%

names related to legal cases or criminal incidents involving perturbation in a neighborhood community

No Comments

No Known Activations