INDEX

Explanations

phrases related to political positions and beliefs

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

çīĪ

-0.79

 snacks

-0.68

 fortun

-0.68

 Tanz

-0.65

Buyable

-0.65

 sights

-0.65

 Skydragon

-0.65

 seiz

-0.64

 Objective

-0.64

 simul

-0.63

POSITIVE LOGITS

Ļ

1.25

¤

1.18

¬

1.17

£

1.15

ª

1.12

¼

1.12

ħ

1.11

Ĵ

1.11

ĸ

1.11

Ķ

1.08

Activations Density 0.322%

phrases related to political positions and beliefs

No Comments

No Known Activations