INDEX

Explanations

words and phrases related to factors contributing to various situations or 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_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

atri

-0.80

byter

-0.79

lance

-0.71

aram

-0.70

scrib

-0.69

scribe

-0.67

ENG

-0.66

lite

-0.66

 Mush

-0.64

OWS

-0.63

POSITIVE LOGITS

 factors

1.13

 influencing

1.08

 determining

1.06

 motivating

1.06

 contributing

0.97

 exacerb

0.96

 factor

0.94

 predis

0.93

fue

0.92

 aggrav

0.91

Activations Density 1.427%

words and phrases related to factors contributing to various situations or events

No Comments

No Known Activations