INDEX

Explanations

phrases related to the concept of understanding, specifically emphasizing the act of comprehending or grasping a situation or information

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

onies

-0.80

hire

-0.71

boro

-0.69

\/\/

-0.69

onto

-0.67

nar

-0.67

gob

-0.66

inating

-0.65

drops

-0.64

iere

-0.64

POSITIVE LOGITS

why

1.13

WHY

1.11

how

1.06

ably

0.91

why

0.91

 comprehension

0.83

 Understanding

0.82

 whats

0.80

 what

0.79

HOW

0.77

Activations Density 0.047%

phrases related to the concept of understanding, specifically emphasizing the act of comprehending or grasping a situation or information

No Comments

No Known Activations