INDEX

Explanations

information related to political figures and events

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

MAS

-0.84

swick

-0.83

MAP

-0.79

ACA

-0.75

Nap

-0.75

STEP

-0.74

Times

-0.72

efe

-0.71

STAT

-0.70

ç«

-0.70

POSITIVE LOGITS

 possible

0.86

 regards

0.85

 practicable

0.83

 contacting

0.71

 outsiders

0.71

 replacements

0.69

 localization

0.69

 respecting

0.68

 anyone

0.68

 interpreting

0.66

Activations Density 0.022%

information related to political figures and events

No Comments

No Known Activations