INDEX

Explanations

phrases containing the word "unknown."

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

©¶æ

-0.91

apers

-0.79

utic

-0.78

boa

-0.76

odcast

-0.72

hibition

-0.71

ushima

-0.71

uilding

-0.70

acket

-0.70

nea

-0.70

POSITIVE LOGITS

 quantity

0.95

 unknown

0.82

 Mortal

0.81

erness

0.78

0.73

 Unknown

0.72

Origin

0.70

variable

0.69

 jurisdiction

0.67

 origin

0.66

Activations Density 0.032%

phrases containing the word "unknown."

No Comments

No Known Activations