INDEX

Explanations

legal terminology and discussion related to law enforcement

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

 vulner

-0.73

 leaps

-0.72

 Drawn

-0.69

 Mirage

-0.67

 conduc

-0.66

 ways

-0.66

 synerg

-0.66

 feats

-0.65

 spir

-0.64

 climbers

-0.64

POSITIVE LOGITS

ï¸ı

1.09

same

1.00

vernment

0.95

\-

0.93

actual

0.91

axis

0.88

0.84

ACP

0.83

#$

0.82

ternity

0.80

Activations Density 0.079%

legal terminology and discussion related to law enforcement

No Comments

No Known Activations