INDEX

Explanations

words related to legal cases, court proceedings, and procedures

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

accompan

-0.67

rss

-0.63

cial

-0.63

iHUD

-0.61

EStreamFrame

-0.61

indust

-0.61

 fare

-0.60

ricular

-0.59

ä¹

-0.59

stem

-0.58

POSITIVE LOGITS

 alike

0.80

 suspicious

0.80

 ashamed

0.79

 suspic

0.79

 bored

0.76

 favorably

0.74

 puzzled

0.72

 disgusted

0.72

 incred

0.71

dag

0.71

Activations Density 0.057%

words related to legal cases, court proceedings, and procedures

No Comments

No Known Activations