INDEX

Explanations

mentions of violent events, political leaders, and unrest in various regions

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_fs3072-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

3,072

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

 darn

-0.84

 Lovecraft

-0.82

Todd

-0.81

Ohio

-0.77

 Whitney

-0.76

 Philly

-0.75

Utah

-0.75

 Gotham

-0.73

 goofy

-0.73

NYC

-0.73

POSITIVE LOGITS

 authorised

0.98

 organis

0.97

 Azerbai

0.96

 manoeuv

0.96

 organise

0.96

 organised

0.94

ÄŁ

0.93

 organisers

0.92

 parliamentary

0.92

 parliament

0.92

Activations Density 6.873%

mentions of violent events, political leaders, and unrest in various regions

No Comments

No Known Activations