INDEX

Explanations

terms related to effectiveness, particularly in fields like law enforcement and healthcare

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_fs12288-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

12,288

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

Hop

-0.69

forest

-0.68

Boh

-0.64

pper

-0.63

umble

-0.62

 Yose

-0.60

 pilgrims

-0.60

ploma

-0.59

bery

-0.59

 Fever

-0.59

POSITIVE LOGITS

 altru

1.01

iveness

0.85

ative

0.81

 deterrent

0.81

å§«

0.77

nesses

0.74

rations

0.73

acies

0.73

 contraception

0.72

cheat

0.72

Activations Density 0.067%

terms related to effectiveness, particularly in fields like law enforcement and healthcare

No Comments

No Known Activations