INDEX

Explanations

information related to geographical locations and political actions

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

 suggestion

-0.79

 Squadron

-0.76

Wing

-0.73

 pedals

-0.71

 suggestions

-0.69

 bribes

-0.66

SpaceEngineers

-0.64

 Razer

-0.64

RTX

-0.63

 advice

-0.63

POSITIVE LOGITS

 borders

1.13

population

1.13

 populous

1.04

 economically

1.02

 annexed

1.01

 coastline

1.01

 industrialized

0.92

 population

0.91

GDP

0.90

 poorest

0.89

Activations Density 4.298%

information related to geographical locations and political actions

No Comments

No Known Activations