INDEX

Explanations

instances of conflicts or confrontations in a given context

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Mechdragon

-0.68

 couch

-0.67

PU

-0.67

 guided

-0.66

 nearest

-0.65

 protected

-0.64

 dust

-0.64

 compliment

-0.61

 exits

-0.60

Adv

-0.59

POSITIVE LOGITS

't

1.71

ÃŃ

1.13

iting

1.02

itely

1.01

eness

0.98

ited

0.98

ned

0.98

Â´

0.97

uts

0.95

cest

0.95

Activations Density 1.128%

instances of conflicts or confrontations in a given context

No Comments

No Known Activations